In one of my recent posts I mentioned the grey cable that is part of our standard mini-sas to sata cables. You can go and find that post if you want to find out more about that one, but …

I pointed out our integration guide to one of our readers to get an idea of how this is used. It reminded me that I often get the question from customers … “the guide only shows 3 series controllers - I have a 5 series controller”. Good point that. However connectivity between the 3 and 5 series controller is the same. If you see a diagram that shows you how to cable up a 31605 to a Supermicro 836TQ backplane then it’s the same process for the 51645.

So go take a look at our integration guide … it’s a very helpful tool.
http://www.adaptec.com/integrationguide/introduction.htm

Ciao
Neil

I was discussing a customer’s configuration the other day and noted, just as a matter of course, that he should update all firmware involved in his system … motherboard, backplane, card and drives and OS driver.

What surprised me, I presume because I’m used to this stuff, was the customers reaction to updating drive firmware. It appears that he has never had to do this before. His reaction was one of complete incredulity. “Surely you’re joking” was the polite version of his response.

Now for the last year this has been common practice for every system I deal with to update drive firmwares. However, it seems that it’s all down to a vendor/product’s credibility status when it comes to updates. If I was talking to the customer about Windows, updates are just a matter of routine. Even when discussing Adaptec RAID cards customers are pretty used to firmware updates (we do them quite a lot). It’s interesting to look at motherboard and backplane firmware updates. The customers are generally aware of them, and might have done them once or twice before, but not commonly.

When it comes to drives, many of the old timers in the industry have never done one, and are pretty scared/nonplussed when told they have to do 16 of them in one server (and that there is no 100% guarantee that the drive’s data will remain intact), and are pretty hard to convince that I’m just not leading them up the garden path (or trying to dodge the blame). Therefore it seems the pecking order for credibility of getting it right the first time is:

Drives
Backplanes
Motherboards
Cards
Operating Systems

As the list goes down people seem more willing to accept that there will be changes and updates are necessary. Interestingly, while typing this little ditty I’ve come to the realisation about how much software is involved in each of the components mentioned. Drives certainly have software, but (and this will get me into trouble) it’s not as complex as an OS. Backplanes, at least in the form of smart backplanes with SES2 processors certainly have software (firmware). Motherboards have lots (BIOS etc), cards even more (BIOS and their own low-level operating system), while OS are, of course, just pure software.

So it would appear that the weak link in the chain is the software component of each product. Now that’s not really telling anyone anything … we all knew that. However, the next time a tech tells you to update a firmware don’t be so surprised … the stuff is everywhere.

Ciao
Neil

I spend a lot of time rabbiting on about using the correct RAID type to suit your data. RAID 5, 6, 50 or 60 (parity RAID) for streaming data and RAID 1, 10, 1E for random data (database). However, it has come to my attention that not a lot of people are really fully aware of exactly what type of data they have.

There is now an easy answer to work this question out. Adaptec are releasing a software product called “MaxIQ Predictor”. The product is designed to capture data from a system, then analyse it to determine whether that system would be a suitable candidate for the MaxIQ Hybrid Array (SSD Caching) Product we now sell, but this new software product actually has many more uses.

Yes, it will tell you whether you’d benefit from SSD Caching, but much more importantly, it will tell you the percentage of random vs sequential reads and writes that are happening in your system. For most people this will highlight the type of data they are running on their systems. This is useful information for determining exactly what sort of RAID array, and what type of disks you should actually be running in your system to achieve best performance (with or without an SSD Cache drive).

This product will be available on our website soon. It’s a command line tool that uses industry standard ways of capturing data that can be analysed by the predictor tool. You don’t neeed to be running an Adaptec product for this to work, or even in fact have RAID in your system … you just capture data for a couple of hours then analyse the output of the system, changing a few variables so that you can set your baseline for your current system. The software currently works on Windows and 32-bit Linux, and I hear on the grapvevine that we’ll be spreading this to VMware and various other platforms in the near future.

The really interesting part is when I’ve run this on customer systems most customers are very surprised at exactly what sort of data they are pushing through their storage … most have no idea.

So if you’d like to try this software now, get a hold of your local Adaptec Field Engineer (don’t ask a salesperson unless you want an earful about our products ). Alternatively, send me an email (neil_cameron@adaptec.com) and I’ll send out the software to you. I can just see my editor (Adaptec Marketing Guru) falling off his chair about now, but what the heck … we have the technology, it’s out and about so why shouldn’t everyone be testing their system to see whether their current RAID type is actually suitable for their data … no matter what or whose RAID product you are using.

I’ll be very interested to see if this blog makes the cold hard light of day.

Ciao
Neil

When building an array using Adaptec RAID cards (hardware RAID 5 that is), you have various build options … clear, build/verify, quick init and (on newer firmwares) skip init.

These options give you choices in how you build your RAID 5 array. While choice is good, and the marketing team love to sprout the line “flexible initialisation options” (because it’s a sexy thing to put on tech sheets), there are issues that users should be aware of when using these options.

As a tech I don’t like giving users all these options. At least not visible on the main screen of your array-build process. Personally I’d like to hide a couple of these options from the user and force them to do things in what I regard is the “correct” manner.

My pet hate is “Quick Init”. Skip Init is even worse but we give plenty of warnings when using this option that in fact you should not be using this option unless the sky is falling and you are talking to an Adaptec Support Professional. But Quick Init is a favourite amongst system builders because it sounds like a good idea.

System builders are always in a hurry. I find this a little frustating as they are generally building a server that is destined for many years of service … and they want reliability, performance and flexibility during those years. Yet they are not willing (in general) to take the time to do things properly in the first place (personal opinion that is bound to slight many builders but what the heck).

Did Michelangelo rush the paining of the Sistine Chapel? No, in fact many Columbines later he took his time and got it right the first time. The result of his initial care was a product that lasted for more than just a few years and has stood the test of time quite well.

So back to building RAID 5 arrays. When you use Quick Init you effectively lay out the array structure on the disk, but do not create any parity across all the stripes. You also, without most people knowing, fix the array into what we call “Full Stripe Writes”. This means that the entire stripe is written in one hit each time anything is written to that stripe. Whether it be a small amount of data or a large sequential write, the whole stripe is written (and subsequently parity calculated for that stripe in the process). This gives a major performance hit for small writes. While RAID 5 is not fantastically good at doing small writes in the first place, it is very, very poor at doing them in full stripe write mode.

We can go into more technical details at a later date if anyone is interested, but the moral behind this story is … use CLEAR or BUILD/VERIFY (clear is my favourite) when building your array. You will take a bit longer to build your system, but like Michelangelo you will create a product that will perform correctly from day one, not run like a dog for your customer. Of course there are many builders out there who either don’t know or don’t care, and just foist the box on the customer as quickly as they can, but for those who are interested in doing things correctly … don’t use Quick Init.

Here endeth the ancient painting (and RAID 5) lesson.

Ciao
Neil

Well, deep down I really do hope you’ve been asking questions, but you probably stopped pretty quickly when you got no reply.

Not surprising since I only just worked out that email questions have been landing in a bit-bucket somewhere that is most certainly not my email address. This, however, I am assured will be fixed “real soon now” (the famous Microsoft RSN). I’ll let you know when our IS department finally track down what is going on here and you can get through to both the Storage Advisors and Linux Blogs.

Not that I’m involved in the Linux blog … that’s a foreign language to me.

Looking forward to a properly functioning blog coming to a web site near you real soon.

Ciao
Neil

It sort of sounded like “See Paris and Die” but now that I’ve typed it I don’t think it works. However … Adaptec have a stand at CeBIT. Why should I be interested? I put my hand up to stand on the stand and talk to people endlessly about our products, but the marketing team knocked me back.

They weren’t convinced that even the english-speakers in the crowd would understand my Australian drawl or unusual language … and seemed convinced that I was just looking for a junket to Hanover in Germany and to build up my frequent flyer points further. I did explain that it’s not October and no self-respeting Australian would be interested in going to Germany at any other time than the “fest”, but that didn’t seem to carry much weight.

Therefore both you and I are gong to have to wait for news on all the latest and greatest products to come out of Europe’s largest technical trade show. Hopefully some lucky journalist writes all about the upcoming products he sees (and sees but is not supposed to see) and gives us something to drool over until Computex in Taiwan later in the year.

At least I stand a chance of getting to Taiwan (and yes, the beer is pretty good there too).

Ciao
Neil

Just had a discussion with a customer who has a 5805 raid card and 7 1tb drives (vendor not an issue here). He originally contacted me asking what the grey ribbon cable on the cables provided with our card were for, but we quickly moved past that to looking at his planned system config.

Note that all drives are 1tb in capacity, and all in a hot-swap backplane. Customer is installing Windows 2008, Exchange (full) and will be using the server as a fileserver as well as an Exchange server. CPU, memory etc were all more than adequate.

Planned spec was:
2 x 1tb drives in RAID 1 mirror for OS (1tb capacity)
4 x 1tb drives in RAID 5 for Exchange and data (3tb capacity)
1 x 1tb drive as hot spare

Problems

1tb for an OS installation is beyond what even Microsoft needs these days, and it wastes a fantastic amount of space. Having 4 drives in a RAID 5 is not an issue, except for the fact that Exchange is a database and works much better on RAID 10 than on RAID 5 … the small writes involved in Exchange are not friendly to RAID 5 (or vice versa).

Capacity wise the customer is ending up with 3tb for Exchange and data.

My suggested config for this server is …

1 x RAID 10 on 6 drives for OS - 100gb capacity … this will use 33gb off each disk
1 x RAID 10 on 6 drives for Exchange - 200gb capacity … this will use 66gb off each disk
1 x RAID 5 on 6 drives for data - based on the fact that 1tb drives are generally around 930gb in capacity there will be approxmiately 830gb left on each drive. Making a RAID 5 from 6 of these drives will give around 4tb capacity.

The remaining drive will be a hot spare.

So …

Customer gets

(a) system OS of correct size … which runs faster than it would on a mirror
(b) a fast RAID 10 for his Exchange … which runs a lot faster than it would on RAID 5
(c) a RAID 5 for data which is 4tb in capacity for data (which is a whole lot better than his current 3tb for both exchange and data)

End result … a lot quicker, better utilised system.

This is a classic example of how “knowing” about the capabilities of your RAID card can give you a much better result than just doing the same old things you’ve been doing for the last 10 years.

So what was the grey cable for? That’s another blog if anyone is interested.

Food for thought.

Ciao
Neil

They’are at it again. Just been talking to another customer with one of the most wasted hardware configurations I’ve ever come across. We regularly talk to gurus in the graphics and media industry who want the fastest possible machine they can get. Some want redundancy, some don’t care (so some use R5 and some use R0).

So the customer has spent oodles on 2 CPU, truckloads of RAM, a video card that would run Avatar, an Adaptec 5805 (good choice) and a bunch of WD Raptor SATA drives.

Sounds like a corker of a system. However, the system builder set it up like this:

2 x 300gb drives in R1 for OS
4 x 300gb drives in R10 for data
2 x 150gb drives in R0 for OS swapfile

Next, the power saving settings:

OS array - spin down after 2 hours inactivity
Data array - spin down after 1 hour inactivity
OS swapfile array - spin down after 30 minutes inactivity

There is some sort of backplane involved, and 3 of the drives are only connecting to the card at 1.5Gb/s rather than 3Gb (haven’t worked out why but I’ll be looking for drive jumpers).

To top it all off, write cache is disabled.

Now why on earth would you do this? I truly struggle to understand how you could build a system like this. The customer runs Windows 7 (some version or other of that wonderful new toy) and spends his time opening and closing 50-80gb files. The system is (a) not running anywhere near like it should and (b) is underdone severely on capacity.

So how would I have built this system …

8 x 300gb drives
1 x 100-150gb R10 for the OS and swapfile
1 x R5 taking up the rest of the disk space

Battery or no battery, ZMCP or no ZMCP, the write cache will be turned on. All drives will be connecting at 3Gb or replaced until they do. Power timers will be the same for both arrays. Probably around the 2 hour mark of inactivity will see the drives spinning down.

The OS will benefit from R10 … it’s relatively random in nature. I actually don’t think the customer will even stress his swapfile because he has a lot (and I mean a lot) of RAM in the system but it will sit nicely on the R10 as well.

The data will sit on the R5. Since the files are large (very large) and are completely sequential in nature, R5 will both read and write quicker than R10 in this scenario. The usable space of this drive will be over 1.5tb, which is a whole lot better than the 600gb the customer currently receives from his current config.

The system will be (a) much simpler and (b) much faster.

It would seem to me that system builders are just not taking customer needs and requirements into consideration when building systems. They are simply plodding along with the same old same old as they have for years, and the customers are suffering because of it.

If this end user wants to really get his machine up and running he is up for a total system rebuild and a couple of new drives. We may settle for OS and swapfile on a 150gb mirror, with a R5 over the 6 300gb drives, but that will be a financial consideration for the customer because he’s already been sold the current disk config.

When you build a system using an Adaptec RAID card, install the management software (Adaptec Storage Manager). Look at the properties of everything and check if all is working well. When you build a RAID 5 array use clear or build/verify … don’t use “quick init” … it’s dodgy and will make the array run like a dog until it’s verified.

But more than anything else … think about what the customer needs and if you don’t know how to build the system, admit it and talk to Adaptec … surprisingly our advice is free.

Ciao
Neil

OK folks, I’m doing a market survey. I think that’s a perfectly acceptable use of this blog (whether my editor agrees or not is another matter).

We sell a lot of 2 series raid cards, specifically the Adaptec 2405. This is a great little card that does RAID 0 (ouch, don’t touch), RAID 1 (mirrors) and RAID 10 (4 drives).

My question to you is … how many people are just using this card with 2 drives? So how about it … tell me how you are using the 2 series card. The best answer will get absolutely nothing in the way of a prize, but some feedback would be greatly appreciated.

Ciao
Neil

My world outside of computer storage is BMX Racing. My kids are very, very good at it and I’m very, very involved in the administration of their sport. So being a computer nerd and a BMX administrator I combine the two and have been developing and delivering training to people who use, or are planning to use, the specialised software package that we score BMX racing with.

I’ll do this backwards because that’s the point of this blurb. At the end of the day, when people are finished their training, they are (almost) always glad that they have participated, invigorated to learn more and view their roles and the technology with much more respect. This in general makes them a much more valuable person to their organisation and an asset to the sport … which was the whole point of the exercise in the first place.

However, and this is the back to front bit … none of them really (a) wanted to do the training, or (b) thought that they would really get very much from it; or (c) were pretty convinced that they knew as much as they needed to and therefore this would be a “waste of time”.

So where does fit with RAID?

If you needed to ask that then you are a definite (c) above. I can also tell you that you don’t know as much as you need to know, that you don’t know as much as is good for you and your organisation and that by learning more about the technology you will become a more valuable asset to yourself and your organisation.

So how and where do you learn more about RAID? Adaptec don’t really run qualified training courses any more (and neither do our competitors as far as I’m aware). So you read a bit here and there, but there’s not much information available in the general computer press as they don’t’ find RAID the most interesting of subjects. They certainly don’t go into technical detail or function of cards … the press (whether it be magazine or online) tend to generally look at the spec sheets from the vendor, or focus on the performance characteristics of the product.

This is somewhat like the motoring journalists of the world when it comes to the Bugatti Veyron. Do they harp on the fact that it develops maximum torque at certain revs so when cornering you are best to be in a gear that leads you into the corner at the bottom of that rev range so you have pull out of the corner? No … they all go “350km/h!!!”. So the reader of their magazines found out what is possible, but not how to do anything of real value.

This is similar to RAID card testers. Yes, you can attach 256 hard drives. Yes, you can put 32 drives in a RAID array. Does anyone tell you the pros and cons of doing something as crazy as this? No, because generally that person doing the testing doesn’t know the answers either.

So where do you go?

Frankly my dear, I don’t give … (Rhett Butler I’m not). Yes, I do care, and no, I don’t have an answer. Tom’s Hardware is about the best of websites that I’ve found for giving technical information along with technical specifications … the “why” that goes along with the where and how, but even they don’t wish to bore their readers to death so don’t dig into a lot of detail of real value to the technician who doesn’t even realise that he needs to learn something, let alone want to learn something.

The reality is that a very small number of people learn a very small amount by talking to people like me. If they don’t fall asleep 3 minutes into the conversation they normally come away with some snippets of information that are of value to them. Problem is that we’re just skimming the surface of people who need to learn something more about the technology they use on a day to day basis.

Is this blog the place to try and teach people? You tell me. Is it the techno-nerds (not an insult) who read this thing or is it the people who want generic information so they can get a general understanding of a technology and then get their techs to drill down and understand the fully story?

Even when I talk to myself I get answers, but bloggers are supposed to write this stuff without care or notice of whether people are reading or taking any notice of what is being written. I certainly am no journalist, and am not the “super engineer” who can answer every question, but just how do we go about getting information out to people in a manner that keeps them awake, keeps them coming back and most importantly benefits themselves and their organisation.

Oh well, more things to ponder but the train is coming into my station and the wife is jabbing me in the ribs so I’ll have to end it here, get off the train and go do some real work in the real world. There, another meaningless blog that did not mention anything technical … I’m getting good at this blogging stuff!

Ciao
Neil

While in China I’ve come across a lot of customers building very large storage systems. Much of this is for video surveillance. They want a big, cheap bucket into which they will record hours and hours of surveillance from millions of cameras (and yes, there are millions of cameras in China).

Now “performance” is never seen as an issue when building these boxes. Capacity and price are the only real requirements. This brings about an issue for me where I totally disagree with my own engineering team. Now I’m supposed to toe the company line here and it will be interesting to see if the editor allows this one to slip through the net, but I have a real problem with the way these large “buckets” can be built.

We currently allow 32 drives to be used in a RAID 5 array. This means 32 x 2tb drives in a single RAID 5 array. The capacity will be somwhere around 28-29tb. Now even on our 3 series card that works fine … when it is running fine that is. However this scares the living daylights out of me when it comes to a RAID rebuild. If one of these drives fails (which means “when”, not “if”), and the system is busy (many cameras recording at the same time), then the load on the system becomes quite ridiculous.

The customer needs to (a) rebuild the array as quickly as possible and (b) not have any disruption to data capture during that time.

We can do the (b) bit … the cards can handle considerable throughput, but the (a) part on this size array is scary. I’m a big fan of multi-level arrays … RAID 50 or 60, or even just RAID 6 if you don’t want to go there. But to put your array at risk of a drive failure during the time it will take to rebuild this particular scenario is very very scary indeed. Especially in the surveillance scenario where backups just don’t happen.

I can’t give you exact figures on rebuild times on these arrays. It depends on speed of RAID card, speed of disks, number of drives in array, load on system etc etc, but you can generalise and say “it will be a long time”. So use some common sense when building these arrays and think about the consequences when things are not working correctly (ie in a failed drive scenario). Make sure your system can cope with the load of rebuilding and capturing at the same time.

At the very minimum have a hot spare in the system so the rebuild kicks off immediately that there is a failure. At best use multi-level arrays (50 or 60) or RAID 6.

Just don’t leave yourself swinging in the breeze any longer than you need to.

Ciao
Neil

Howdy folks,

No, I haven’t disappeared off the the face of the planet … currently in China extolling the virtues of all things Adaptec to those who will listen or can understand me (no, I don’t speak Mandarin apart from a few common words). Thought it was time to do something about this blog for 2010 in a quick break between meetings.

So, happy new year, hope you all had a good break, etc etc etc.

China is going “green” in a big way. Power reduction is a major focus in this country, much more than I’ve seen in any other country where I travel. Yes, there is a factor of saving money, but also a major emphasis on reducing carbon footprint (it’s a national issue here).

However … as usual I’m finding not many people understand the savings to be made by not running your hard drives in your server 24×7. The ability of RAID cards to stop spinning drives after a period of inactivity is not something widely known or understood, but is viewed with great excitement when it is explained fully.

Just like many countries, there are several periods of the year when servers are not used … the best example is CNY (Chinese New Year) coming to a venue near you real soon. During that period many organisations will be closed for a minimum of two weeks. That’s not to say that people won’t be still working from home, checking emails occasionally or logging in remotely to their servers, but the general rule is that a lot of servers will be idle for a period of time during this holiday.

Now of course they could just turn their servers off for this period, but who does that? Even for the Christmas break in western culture I’ll bet that most companies just left their servers running, with all their hard drives spinning merrily away chewing up electricity, heating the room and causing their cooling systems to run full pelt (especially in good old boiling hot Australia).

What a waste.

I’d like to know just how many people actually utilised the free features in their RAID card to reduce their running costs over the Christmas break (and the coming CNY). I’ll bet it’s not as many as could have. Now the system admin lads just counter this with … I don’t pay the electricity bill … that’s the account departments problem. But in reality you can save your company a lot of money without affecting the performance or functionality of your systems.

So how about it … who out there is “Green” (using power saving that is)?

Ciao
Neil

Well folks, it’s that time of year again … eat too much … drink as much as the wife will allow and generally forget about all things computing for far too short a time.

For many of us this is time to chill out, hit the beach, sink a few tinnies and generally recharge the batteries. However for some it’s pressure cooker time. There is many a server rebuild and upgrade happening during the break, with many a system admin sweating on the fact that the server will be up and running and all working correctly for the start of the new year.

So here’s my simple tip for less stress. Image the server before doing anything. I always find that if I have taken all the right precautions and made the proper backups, the stress is less and things generally go well. If of course you’re swinging by the seat of your pants with no backups in place things are destined, fated and bound to go wrong.

Adaptec tech support will be around for a lot of the break and you can always ring Australia, Europe or the US if you can’t find tech support in a suitable timeframe or no-one seems to be answering the phone.

To everyone who has contributed throughout the year I’d like to thank you for taking the time to read my ramblings and (I know you’ll love this one) … I’ve been promised some assistance in the new year to make this somewhat more interesting. Mind you I’m not holding my breath but you never know.

So thanks for 2009, have a great Christmas and New Year and let’s look forward to some better economic news and a bit more enthusiasm in 2010.

Ciao
Neil

Just finished a quick road trip to the other side of the country where I was espousing the benefits of our new MaxIQ product. As a consequence of talking to database integrators we spent a lot of time discussing existing implementations that they were having problems with (and therefore candidates for MaxIQ).

Something that came to light on a regular basis was the fact that a lot of integrators use RAID 5 for every system they implement, whether it be a fileserver or a database server. Now RAID 5 is a good all rounder. It’s great for fileserving and general server use, makes good use of the available disk space and most people are comfortable enough with the technology to actually think they understand it.

So what’s wrong with that. Simply put, RAID 5 is (in general) no good for environments with small random writes. Since I was promoting MaxIQ which is excellent for small random reads, naturally I found myself in an environment where there were a lot of random writes. On almost all occasions customers were using RAID 5. Most were using SAS disks, which meant they had recognised the signfiicance of the performance issues they were faced on database servers and had opted to offset their performance issues with fast-spinning/seeking disks.

Therefore we had a scenario where customers were trying to fix performance problems with hardware alone. Put a faster RAID card in the machine, put faster disks in the machine, add more RAM, improve the processor … but what about something as simple as using a different RAID level? RAID 5 is a great performer on many disk types, over a wide variety of read/write scenarios and data sizes, but it has one weakness. Random writes become slow because of the minor stripe write characteristics of RAID 5. There are multiple reads and writes on disks, plus parity calculations to be made for every small write in a RAID 5.

So what to do? Simply for most database applications you should consider using RAID 10. RAID 10 does not do parity, but simply writes the same data in two separate disks within the array. Consequently for a variety of technical reasons RAID 10 has faster random writes than RAID 5. Yes, there are scenarios where RAID 5 is good for database (in mostly read-type database environments), but in general, especially in the SMB market with accounting database, SQL, Exchange etc, RAID 10 is a better option for database performance.

Remember … you don’t have to make your entire disk structure of the same RAID type (mix and match for your different data types), you can have more than 4 drives in the RAID 10 and you can have different disk types connected to the same card, so now you can mix your raid type across different physical hardware arrays (just don’t put both disk types in the same array).

Point of the exercise? Hardware alone won’t give the performance every time. It will help, but you need to keep an eye on your RAID type to ensure it matches your data set.

Ciao
Neil

I was working the stand on a tradeshow recently when a very nice gentleman stopped by to ask me some performance questions about our RAID cards. This is not an unusual scenario, but this customers system was pretty unique and it made me think quite a bit about what different poeple are looking for from a RAID card.

The customer deals with video editing, so performance, performance and performance are important. Did I mention performance? That’s all the customer is interested in. Redundancy didn’t come into his mind (until I pointed out what a pain reinstalling Windows was). Now this customer had spent quite a bit of money. Dual quad-core Xeon processors, 24Gb of RAM, endless video card power, a 5805 Adaptec RAID card and 8 x 300gb WD Velociraptors in a Supermicro hot-swap server chassis. All this for a high end graphics machine was, I thought, pretty crazy stuff. However, the customer was less than impressed with the storage performance of the system.

Investigation showed that the system builder (remaining nameless) had set 2 drives in a mirror for the OS, 2 drives in a RAID 0 for the first data volume, 2 drives in a RAID 0 for the second data volume and 2 drives in a RAID 0 for the third data volume. This config had given the customer the separation of OS from data he was looking for, and the three data volumes he had requested, but it simply stifled the performance of the system, along with wasting a fair amount of space of expensive drives.

After a long discussion with the cusotmer on exactly what his needs were, I gleaned that the OS and software installation was in fact quite a time-consuming operation, and the he really wanted protection for that part of his system, but 300Gb was a waste of space. It also is, in my opinion, not a very fast way of running an OS. As for the data volumes, the customer was just plain interested in speed, speed, and a bit more speed. He did not in any way care if a drive failed and he lost all his data … he wanted to be able to rebuild the system very quickly and copy his data back on from his video sources. So RAID 0 was perfect (never thought I’d hear myself say that).

So we reconfigured his arrays. Starting from scratch we created a RAID 10 over all 8 drives of 100Gb capacity. This was more than sufficient space for OS, software, swapfiles and miscellaneous stuff, and gave much higher performance than the 2 drives in a mirror previously configured. For his data we configured 1 large RAID 0 of all the space that was left on the drives. Creating the unique partitions he required was done at the OS level. End result of the config was that the system appeared to all intents and purposes to be the same as the system he had purchased.

However, now performance absolutely rocked. The customer was gobsmacked at the improvement in speed. Why did it improve so much? Because now every read and write involved (potentially) 8 drive spindles instead of a maximum of 2 at any time in the previous configuration. Yes, there is danger in losing a drive and losting the RAID 0 and its accompanying 3 OS partitions, but the customer realised he could put that back to together in only a few minutes after replacing the drive (with the cold spare he already had).

Result: one very happy customer.

So where’s the point to this ramble? I cringe when I hear RAID 0 … it’s dangerous to data and I would never normally recommend it. But after listening to this customer and questioning him about his needs and reasons, the final system build made perfect sense. Just because it seems a little crazy to me doesn’t mean its crazy, just different to the day to day email and blogs I spend my life writing.

I just thought this a good example of how listening to your customer, and focussing on his needs rather than your fears, lead to a happy customer, which is much more important than having a comfortable technical consultant.

Ciao
Neil

This should have nothing to do with RAID. I should not even ever have to comment on such an issue. You’d think in this day and age with all our fancy technology we’d have this right by now … NOT A CHANCE FOLKS!

Chatting with the tech support folks over a few ales is like watching an apocalyptic movie … you know the kind where the world ends in the next ten minutes with huges waves and crashing thunder etc etc. Except in this case the waves are the tears of workers who have lost their last three weeks work and the thunder is the sound of the door slamming behind the ex IT person as the boss boots him/her out the door.

Whenever anyone rings Adaptec for help, we always ask them: “what state are your backups in?”. This generally gives us an idea of whether we are dealing with a professional, organised organisation or otherwise. More often than not its otherwise.

Yes, we still hear the old chestnut “I don’t need backups … I’m running RAID 5″. OK, have a laugh. Think to yourself “he’s making this up”, then think again. There are still people out there (my editor won’t let me describe them accurately) who don’t do proper backups.

Almost everyone has good intentions. Everyone has a backup of some description lying around the place. Most people have a backup that is at least a month old they can “probably” restore. The problem lies in checking your backup and restore processes. Try this quick quiz … delete (rename to something you can remember) a document that is vital to your business operation (ie your accounts database). Then sit back with the stop watch (or generally a sundial will do for this one) and see how long it takes the IT gurus to get your workers back working again.

Then do some rough calculations on what those minutes, hours, days, weeks cost your organisation. If I asked you to give me $10,000 out of the goodness of your heart you’d think I’m slighly bonkers. If you lose your server in even a medium sized organisation for just 1 day you’d come close to losing the same amount of money. So why not just give the money to me and get your backups sorted.

RAID won’t save your bacon every time. It’s designed to save your data from disk failures … not viruses, corrupt operating systems and/or less than intelligent users (of which there seem to be plenty out there). And sometimes, yes just sometimes things even go wrong at RAID level and you’ll need to rebuild your system.

(Edit: What brought this rant on? A customer who had a total system crash/failure/calamity in the middle of a RAID rebuild on 4 old SCSI drives whose backups were 3 weeks old. Got the news today that we saved his data and he was able to retrieve it all after following a very long, tedious and stressful process I gave him. So while he didn’t lose his data he lost a lot of sleep and several years due to stress … all of which could have been avoided.)

So how good are your backups?

Ciao
Neil

I recently read a draft document aimed at outlining RAID maintenance “best practices”. This has been developed to counter some of the insane practices of our customers such as recycling used drives in RAID configs because the drive “seems OK”.

While the document jumps pretty much straight into proprietary and unique Adaptec features, it made me think of some fairly generic steps a customer can do to protect their data that fit almost every scenario or vendor.

Management software - install it. It’s amazing to me the number of servers out there that don’t have RAID management software installed. Installing the software is the first step … looking at it on a regular basis is the next step. Yes, you can set up all sorts of alarms and notifications in all manner of software, but I certainly have been caught by the network gurus blocking ports on systems I set up ages ago, and while I’ve set the system up correctly, its not working now because of other factors. Simply running the software and actually looking at it is half the battle.

Physical monitoring … wow, I’ve never seen that red light on the front of the box before! Looks pretty doesn’t it? Again, all the software in the world doesn’t replace a set of eyes and some grey matter between the ears.

Email and system notification. Set it up. Test it. Check it regularly. Will your system actually send you a message when something fails? Is the email address it is sending information to still relevant or has that person left the company and that Exchange mailbox been closed? One of the major causes of total RAID failure is customers ignoring an initial drive failure. In a RAID 5 environment they always find out about the second drive failure because the system falls on its face.

Log checking … it’s always an interesting exercise to look in Windows event viewer and find 30 pages of red icons … basically the server screaming about a problem that no-one is taking any notice of. This comes back to running your eyes over your server … physically and via the management software (RAID management softwar and system management software).

RAID integrety checking … all vendors will have a way of checking the integrity of their RAID array on a regular or scheduled basis (or at least they should have). This level of checking is concerned with the consistency and accuracy of the parity data spread amongst the disks. It should be checked on at least a weekly basis (background and automated is the way to go).

Correct drive choice … this one is controversial. Many, many users are running “desktop” drives in servers these days because of price and capacity issues. It’s the drive vendors who specify the difference between desktop and enterprise drives, and you can always get an argument from any tech by stating that one type is better than the other, but make sure you are using drives that your disk drive vendor supports in a RAID environment.

Documentation … what parameters were used when building the array in the first place. What firmware revision was on the card when the array was originally (or subsequently) created? What stripe size was used? Almost every tells me “I just used the defaults”, but almost no-one can tell me what they are . What size arrays did you make (down to granular size details)? All this information and more may seem trivial and easy to do when you are setting up a system, but when the pressure is on in a failure situation it can make life a lot easier for the tech to sort out your issues.

Correct RAID choice … this is often seen as a performance issue (which it is), but it’s also a redundancy/safety issue. If you are using large sATA drives use RAID 6 instead of RAID 5 … the performance difference is almost negligible these days and the safety of 6 over 5 is important.

Up to date software … keep your firmware, drivers and management software up to date. Most customers are always updating their OS, and it makes good sense to keep your hardware, drivers and RAID management software up to date.

Backups … did I mention that you still need backups? I choke when I hear the statement “we don’t trust our backups” or “we can’t be sure of the integrity of our backs”. For goodness sake, this is one you definitely, positively and absolutely must get right.

If you are doing the majority of the above then you are halfway towards have a stable long-term server that will look after your data for you. So how good are you at “best practice”?

Ciao
Neil

I have been seeing a lot of people using 1.5Tb and greater hard drives lately, and find myself getting pretty uncomfortable about

them using plain old RAID5 for 12-16 1.5Tb drives.

With the failure rate of SATA drives not being something to smile about, and the length of time required to rebuild arrays of this

size, I am strongly inclined to recommend RAID6 to users. These days the performance hit is not that great (almost negligible) but

the safety factor of being able to survive 2 simultaneous hard drive failures, or a second drive failure during a single drive

failure rebuild, is, I think, worth it.

However … as drives get larger, the capacity hit of RAID6 becomes greater. “I don’t want to lose 1.5Tb capacity” is what I often

hear from people when I suggest RAID6. Simply put, the increased size of the drive means increased lost capacity when moving from

RAID5 to RAID6. While I think the simple solution is to just purchase more drives, it seems not everyone has my deep pockets

(unsurprising as that may be in these economic times).

But … I’d strongly urge people to consider RAID6 when using large SATA hard drives to cover themselves (and their data) - and be

damned with the capacity issues. I’d rather keep 14×1.5Tb data than lose 15×1.5Tb data.

Food for thought.

Ciao
Neil

I was madly making RAID arrays the other day to do some testing, when a message popped up on the screen … “The selected configuration allows for the creation of a logical device with Enclosure Level Redundancy. This will override any second-level devices selection that you have made. Do you want to configure Enclosure Level Redundancy? Y/N”.

Now I have a bad habit of just ignoring pop-ups (which causes me some pain occasionally), but this one had me
intrigued. Either I was asleep during some training session (not uncommon) or someone left me out of the loop. So what does this mean (the message, not the sleeping bit). I said yes and nothing exciting happened (very disappointing). That really got me intrigued so I looked at the properties of the array I had created.

My old Supermicro 815TQ is an 8-drive 2U system. Even though it looks like one backplane it’s actually considered by Supermicro (and our cards) as two backplanes … a row of 4 drives above a row of 4 drives in separate backplanes. Since I was making RAID 10s using all 8 drives, the card saw something I had not considered (and did not have control of anyway) … that it could make each pair of mirrors in my RAID 10 with one drive on each backplane for each mirror.

Simply put, if one backplane fell over, the system would keep running. The card is smart enough to see an opportunity to add an extra level of security, simply because of the configuration of my system and the raid level I was using. Cool!

The morals to the story are many … read pop-ups, stay awake during engineering briefings and when prompted to do “enclosure level redundancy” … do it.

Ciao
Neil

All the hype in Australia at the moment is the National Broadband Network … basically fibre to the node everywhere (fibre to all homes and businesses). Now no-one believes it’s going to happen in the next 6 months but the current government has a firm commitment to get this going.

So what implications will that have for users?

As a venerable road warrior I live out of my laptop. Like most mobile workers I have everything I need in the laptop, and backup regularly when I get near a link that’s fast enough not to bore me to death. Currently that means LAN. ADSL, which is the most common broadband technology in this country is nowhere near efficient enough to pump large amounts of data across in a short period of time.

But what if the internet link I’m using wherever I am in the country runs at LAN speeds?

I could work from home with corporate storage access speeds that rival my current LAN access speeds when I occasionally make it to the office. All my colleagues could also work from home. Organisations that have offices in multiple locations across the country would benefit dramatically. Imagine having only one server (or cluster) for your entire organisation, rather than a complete server environment (including abackups etc) in each city. Now that will impact storage dramatically.

Cloud computing should benefit dramatically as well. Instead of having that server in the one organisation head office, it could actually reside on a service-providers network somewhere in the cloud, mean every user, whether a road warrior or office-bound worker, would have high-speed access to their data through the broadband network.

Fast data access, consistent and reliable backups, common experience for all users no matter of location (in or out of the current corporate network), centralised storage and organisation of data, easy sharing of data across corporate employees … it all sounds to good to be true.

So what impact will it have on storage? Larger, faster, centralised repositories of data, greater need for tiered heirarchy of data, increased reliance on disk2disk backups as a wider time-range of people access the same data (reducing backup windows) … the implications are many and varied.

For the moment all I can really be sure of is that there’s going to be plenty of work for people willing to dig trenches and lay cables for a few years to come. Now thats a thought … the wife is always telling me I need to get fitter

As my daughter says … bring it on!

Ciao
Neil

I’ve been getting a lot of requests lately from people putting together massive storage “buckets” … 16 or more 1.5 or 2tb drives in a single array to provide a large storage container for whatever their data is (they don’t normally tell me that). However, they all so … “performance is not an issue”. This statement, IMHO, is basically a load of *^(*&%*&$%. Everyone is interested in performance … they just might not realise it straight away.

The question of which card to use often arises in these conversations. The 3 series card looks good because it’s cheap with lots of connectors … and price is a big consideration for these data “buckets”. So we discuss the various merits of 3 vs 5 series. “No, don’t need the performance of the 5 series” is a common statement. But what about, I ask, when things are not going so well.

The 3 series card is fine when your array is stable, optimal and all the planets are aligned, but what about rebuild times? This is where the 5 series with its extra grunt comes in handy. Rebuild times, impact on performance during a rebuild etc are questions that a lot of people don’t think about when building a system. In the past if you had a fairly small SCSI system it didn’t matter too much … the system could rebuild in the background relatively quickly without much impact on customer performance, but that’s not the case today.

When using large numbers of large drives rebuild time is an important factor. When using large numbers of sata drives impact on performance during a rebuild is also an important factor. So don’t think about specifying the products for your system for just the good times. Think about the bad times as well as make sure you put enough performance under the hood to (a) get you back up and running as quickly as possible and (b) not annoy the living daylights out of your users while this is happening.

Food for thought.

Ciao
Neil

I’m sitting here in Hong Kong airport with less than half an hour to knock over some emails and wake this blog up again (or myself, whichever comes first). Why only half and hour? Is the flight home leaving soon? No. As a matter of fact I have another three hours sitting in this bar to while away my life.

The half-hour limit comes from the battery life on the old Lenovo. Great little laptop, but after a couple of years the battery has had the fritz. Now why is this surprising? It’s not, and that’s just the point. Everyone tells me that since I live in front of this screen it’s not surprising that my battery has died after a couple of years use. Why not just plug it in and charge it for a while? Simply because I’ve been in China where the power connectors are the same as Australia but in Hong Kong (even though it’s part of China), the power adapters are a hangover from Britain and I didn’t pack that adapter.

It’s perfectly normal and acceptable to everyone who owns one. Same goes for the mobile phone. I actually don’t even get two years out of a battery on that thing (which will be very interesting to see how long the new iphone actually lasts since apparently I can’t change the somewhat disappointing battery on that one).

So am I just waffling until the little yellow triangle tells me to go back to the Tsing Tao? No. I’ve spent a week rolling around Asia promoting various Adaptec products (which is what they pay me to do), and when I tell people that batteries on RAID cards don’t last the same life as the card … they all laugh at me.

When was the last time we changed a battery they ask? When was the last time we checked the battery? When in fact was the last time we took any notice of the battery? It’s part of the system and is fine. What’s this rubbish about batteries not lasting?

Well guess what. Batteries don’t last. Just like the mobile, or the laptop, or the (… fill in your own electronic device here), the battery on the RAID card doesn’t last.

So why did I entitle this “relativity”? Was it because I just saw a nodding little Einstein on the movie on the last flight (which was quite funny)? No, it’s because I’m always amazed that people (who are sometimes relatives of mine), have relatively naïve ideas on the relative features and capacities of two quite similar products, even when they are in completely different environments.

So guess what. Batteries don’t last. End of story.
The dreaded triangle has arrived so I’m back to the beer.

Ciao
Neil

So the wife was right after all these years (I can say that here because I can definitely, exactly and almost confidently say that (a) she doesn’t read this and (b) doesn’t listen to a word I say about “blogs” and other computer related stuff).

http://searchstorage.techtarget.com.au/articles/35314-EMC-Cisco-accused-of-FUDfest-Apple-hammered-NetApp-accused-of-PR-BS-as-IBM-and-HDS-fight
(you’ll have to read down the page to get my point)

There is nothing that makes you think on your feet faster than answering questions from a journo. So what’s the future of RAID? Is it RAID 4 as per my colleague from NetApps? Don’t think so. NetApps are one of the few people doing it and if you ask joe public he probably hasn’t even heard of it.

6 and 10 are my bet. RAID 6 is the go these days because of faster processing power on RAID cards … the performance hit we once took for running so much parity has pretty much gone and yes, you can survive two disk failures at once.

RAID 10 is also a favourite amongst the paranoid of us out there because it “sounds” very safe, provides great performance and doesn’t cost the earth these days.

There is, however, a difference between 6 and 10 that needs to be understood (and often isn’t in my experience). Both 6 and 10 can survive two simultaneous drive failures … but with 10 you have to be very, very lucky as to which drives fail to survive the two drive calamity. 6 doesn’t care … two drives go and you’re still good. With 10 if two drives from the one raid leg go, you’re gone.

So don’t think about RAID 10 as being able to survive two drive failures … just think about it as a very fast way to use up a lot of disk space and money at the same time.

Ciao
Neil

One x 2.5″ device (laptop thickness) which is a drive carrier (not an actual drive).
This device needs to be able to carry 2 x 1.8″ 320gb SSD drives in RAID1 configuration.

When device is developed please send to Adaptec … I’ll evaluate it in my laptop for a very long time.

Obviously just gone through the pain of a rebuild on the laptop due to hard drive failure. When, oh when, are we going to get RAID in these things?

So all I need is someone to develop this little RAID carrier for me, the drive companies to get the 1.8″ drives up to 320gb, a vendor to sell the whole thing for the same price as I currently purchase a 320gb platter drive and the world will be sweet.

Not asking too much am I?

Ciao
Neil

I’m noticing a slow but consistent change in people’s attitudes towards RAID types lately. It mostly has to do with the price and size of SATA disks. It seems that RAID5 is just not considered necessary, relevant or even suitable these days.

In days gone by RAID10 was always available, but considered way too expensive for anyone but the database boffins who were only interested in perforance at all costs. RAID5 made a lot of sense in servers where a mix of capacity and performance were paramount (especially in the fileserver market).

However it seems that with the advent of large SATA disks the requirement for RAID5’s better utilisation of disk capacity is not such an important issue, and that customers are willing to put together larger RAID10 configurations to meet their capacty requirements.

I personally tend to agree because of the general all-round performance benefits of running RAID10 over RAID5, but I’m interested in what you are doing out at the coalface. Are you still living with RAID5 (and loving it), or has it gone the way of the dodo? (Small caveat here … I’m investigating the performance benefits of SQL on RAID 5 at the moment so might have some different opinions in the future.)

Love to hear your thoughts on this one?

Ciao
Neil

The wife hit the roof this morning. Literally. From July 1 the price I pay for electricity increased by 20% … ouch. This is a significant increase in the running cost of a home and a business. It’s a wake-up call for me to get serious about looking at exactly what uses electricity in both my home and office … and to look at what I can do to reduce the amount of electricity I use.

There are several basic electricity uses in the office that I can work around … reducing the number of lights that are on in areas not heavily used … putting on a jacket instead of pumping the air-conditioning flat out in the cold weather (yes, it’s cold where I live), but the biggest electricity use in the whole office is my lab.

It’s really convenient to have all my servers and NAS devices humming away so that I can access whatever I want, or test whatever I need, without waiting. It’s also necessary for some of my machines to be running all the time whether I like it or not.

So on those machines I turn on power saving on the hard drives. That’s a pretty simple process but let’s look at exactly what it does for me. Now if you want financial number crunching then go to our website … there’s a fancy calculator on there (www.adaptec.com/greenpower) that will let you come up with all sorts of numbers.

However, in simplistic terms, this is the scenario in my home office/lab. I work a ridiculous number of hours (that one is for if the boss reads this blog), but I don’t access my servers all the time. I sleep occasionally (when I don’t access my servers at all) and I do not, repeat do not, work weekends (except when travelling). Now I’ve done a few calculations that are a little scary, but eye-opening:

Hours in a week - 24×7=168
Hours worked per day - 15
Days worked per week - 5
Hours worked per week (maximum) - 75
Percentage of time that I could possibly, remotely, conceivably access my servers - 44.64%

Simply put … that’s 55.36% of the week that I don’t access my servers. Now consider the fact that the hard drives in my servers are all spinning all the time and you start to ask some serious questions … like “why?”

Since, for some reason known only to me, I use Adaptec controllers in my servers, I have enabled power saving on all my raid arrays. This means that the server is still running, and I can access it whenever I want, but the hard drives are not spinning unless I either need to write data to them or read something from them. So in theory I’m now saving 50% of my electricity costs. That’s probably not realistic in several ways. Firstly the rest of the server is still running (power supply, processor and fans etc so that will reduce the savings somewhat. On the other hand, servers like my backup server which only run for less than 1 hour a day gain massive power savings from having the hard drives asleep for a large portion of the day.

I’ll do the maths over the coming months, and watch the electricity bill (as will the wife I’m sure), but it makes sense for even a small operation like mine to save runing costs. Imagine what it can do for a larger organisation!

Ciao
Neil

Once upon a time I coined a phrase “SQL = SAS”. Our marketing department dismissed it as not having enough words in it to keep them all employed so it was relegated to a printout on my wall of ideas.

However … I think I may now have to agree with our marketing department (damn that hurts me to say) … SAS is under threat from SSD (SLC specifically). Following is a link to an excellent article from Tom’s Hardware - written by a journo with skills that make me look like a clay tablet chiseler.

http://www.tomshardware.com/reviews/x25-e-ssd-performance,2365.html

Now the bent of that article is speed in Mb/s (Megabytes per second to clarify this always contentious abbreviation). Ironically that’s not what database is measured or mentioned in (iops are the go with database because of the randomness of the data), in any case over 2000Mb/s is damned fast, and the streaming video lads will be drooling as they read. Their only real difficulty in life is the ridiculous cost of putting together something large enough for their needs.

Database (and here I’m talking SQL) on the other hand doesn’t need the space requirements of video and I’ve always found SQL system builders will go to just about any length (and cost) to improve the speed of their systems.

So take a look at the attached article, specifically on page 7 (database) and you’ll see what I’m talking about. The speed for database is amazing. Would you do this for a fileserver? If you have more money than sense, possibly, but I seriously doubt it. Would you do it for video? Possibly if your budget is massive and your requirements extreme. Would you do it for database? You bet there are people out there who after reading this article will be racing out to purchase some seriously fast solid state drives.

While on this point, go and read my previous blogs about making sure that when you purchase a solid state drive you (a) know what you are purchasing and (b) get the type of drive that suits your needs. Remember there are a lot of variations out there now in the drive market, and and not all apples are in fact apples.

For a bit of fun read some of the comments on Tom’s site (at the bottom of each page). Most people are concerned with how fast this extreme array will boot windows and load applications. Now I’m concerned about that too, but just can’t see myself putting together such a system at home or on a workstation!

Ciao
Neil

Blogs will be a bit thin on the ground for the next couple of weeks (more so than usual that is) as I wander off on a sojourn across Asia. Singapore, India, Thailand … sounds exotic (so the wife thinks), sounds expensive (so the boss thinks) … and sounds like hard work (that’s what I think).

I’m loooking forward to a first hand view of the state of the economies in those regions, what their focus is on storage over the coming year and of course some of the fantastic food across all three countries. A nice sideline to the whole affair is tasting a few ales across the region, culminating with my favourite … the Kingfisher of India.

The trip finishes off with the BMX World Championships in Adelaide, Australia. No, I don’t ride, but the kids are pretty damned good and I’m chief pit crew for M&D Racing. With any luck we’ll have a new blogger joining the ranks soon which will relieve you from having to listen to boring-old-me.

Sounds like an interesting start to the new financial year.

Ciao
Neil

Adaptec released a new product today/yesterday (depending where live and how late it is) … and of course I’m not going to sit here and blatantly advertise it … that would not be bloggish of me. You can find it all over the web (once you work out how to set Google to find only posts in the last 24 hours).

However … in the last 2 hours of trolling the web to see if our marketing team really do work as hard as they say (yes Scott, we’re watching) … I’ve seen some pretty crazy feedback on a few websites. Pretty much some people just didn’t get it.

As much as I dislike turning this into a tech support forum, there is obviously a need for a bit of discussion on this new stuff, so ask away …

Ciao
Neil

Technologically speaking that is (though it’s a scary thing to say in this day and age). It’s pretty bizarre when I find a “blog” is now an outdated form of communication.

If you go to the Adaptec home page, down on the bottom right hand side is the “Follow Us” tab. Inside you’ll find Facebook, Twitter and THEN the storage blogs. I supposed its a sign of the times that I’ve been usurped by something developed by someone who wasn’t even born when I learned to type.

About the only saving grace of all of these technologies is that (a) I can now ask my kids for help rather than them asking me to do their homework and (b) all this digital “stuff” (I can’t bring myself to call it something real) needs to be stored somewhere.

Since we’re in the storage industry I suppose that’s a good thing. Therefore keep twibbling, blottering, clogging, facepainting or whatever you call it … it’s good for business (even if I don’t understand it).

Now where did I put that chisel and hammer … I need to write some more.

Ciao
Neil

Off tomorrow to do another trade show … the usual tyre-kickers intermingled with a very intelligent crowd of business people interested in finding out where the future of storage lies.

Trouble is, of course, it depends on what magazine you read, or analyst you listen to, as to what the future holds.

Personally speaking, I’m finding a lot of interest in SAN. It’s a quiet revolution being driven by the Virtual Storage Crowd. If you are gong to virtualise your servers then you’d better put all their storage on a box you can expand, manipulate, mirror etc … not just in direct-attached drives. The fibre market is still prohibitive in cost for the SMB, but iSCSI helps to bring that back to reality.

Of course, there is a big market for direct-attached even in the virtual server world. There are a lot of smaller users who want to run just a few servers on the one piece of hardware and direct-attached storage is fine for them. But as the number of virtual and physical servers increases, and the complexity of the virtual world and server infrastructure grows, so does the pressure to move the storage for VM’s, data and virtual desktop images onto SAN storage where you have better flexibility.

So at the moment it appears it’s the larger end of town that is interested in SAN, with the smaller side of things sticking with direct-attached storage. But for how long? Send me your thoughts on where your organisations are going with SAN vs DAS.

Ciao
Neil

In a spin about not spinning …

I hope people appreciate how long it takes me to come up with these snappy headlines.

Just spent a few days at Computex in Taiwan … wading through a mountain of new equipment, motherboards, memory modules and especially, importantly and all-pervasively … Solid State Drives (SSDs).

They were everywhere. It seems that there are thousands of these things on the market, ranging from well-known to never heard of. Of course, they are flavour of the month and have a very large target audience, but it was amazing to see how many brands are on the market.

Now my focus is on storage, especially in the server arena. While I would have loved someone to give me one of these babies for my laptop that didn’t seem to happen, so I whiled away my time asking questions of these manufacturers. Now to be fair, there were a lot of very pretty young ladies trying to sell product whom I don’t think were product experts, but there were also the supposed “product experts” hanging around the back of the stand trying not to be bored to death by several hundred thousand tyre-kickers, so there was knowledge there if you asked hard enough questions.

Now what sort of questions would be “hard” questions for the SSD industry? Simply (a) are you using SLC or MLC technology, (b) what are your performance figures and (c) how long do you expect this drive to last in my laptop and an average database server (maybe a 50-user SQL environment).

Question A generally drew some blank looks, but reading the tech specs found the answer (which was, to a very large degree, MLC). Question B was easy … everyone has amazing figures on display (take all with a dose of salt) so there’s no shortage of amazing claims out there.

Question C had people running for cover. “Oh, we are not targetting the server market” was the well-practiced standard response by most manufacturer’s representatives. However when we got to discussing their view of the future drive market, all seemed of the opinion that SSD would certainly take over the SAS market, and that SATA would come under threat as the size of SSD increases and the price comes down.

Great, so I want to build a database server. Yesterday I would have used 4 x 15K RPM SAS drives in a RAID 10 configuration to get my best combination of read and write performance (especially on small writes) and pretty-well-sorted reliability. Today the “new age” drive industry would have me using 4 x SSD drives in the same config with amazing IOP and throughput figures.

So I build my system with SSD … how long will it last? I am aware of the long write-life of SLC technology, but it also has an amazing cost involved, so many people will not use those drives, but drop for what seem a better bargain (which happen to be MLC drives). Now I don’t believe MLC has anywhere near the write life of SLC, so exactly how long will this server last before things start going terribly wrong with my storage.

With my SAS investment I could easily expect 3 years, and more likely 5 years good use (on average) from my expensive, hot, expensive to run and heavy SAS drives, but what will MLC give me? That one I could not get an answer for.

Now the SSD drive experts and manufacturers out there will read this and immediately start yelling at me that I’m using the drive out of it’s intended target usage pattern, and that I should be using SLC technology, however these are the same people telling customers that they should use SSD for everything and omitting the relatively important information about drive life when selling the product and making amazing claims to customers about size and speed, glossing over the underlying technology questions.

Don’t get me wrong … I think SSD is great, it’s here to stay and I want one sooner rather than later, but there are a lot of variables out there with these drives, and Joe Public doesn’t seem to have much of a clue.

It’s either “buyer beware” or do your research first.

Ciao
Neil

I was just in the middle of writing an article on “clouds” when my article disappeared into one (and of course I’m such an expert on this software that I have no idea how to retrieve it). Therefore I’ll just start again in slightly less cheery state of mind …

Reading my local computer rag “Computerworld” I came across the following article: “Gartner: Aussies to shun cloud computing this year … Pragmatic attitude of local IT managers cuts through the marketing hype”. (So yes, now you know where I reside.)

There’s a corker of a line in here: “The cloud name give a sense that anything will apply to it … it can be considered a large amorphous concept, but to an Aussie that’s not useful.”

Not useful? If you can’t cook a Shrimp on it or crack a tinnie with it then of course it’s not useful. The article goes on to state “About 50% of businesses said they will increase investment in virtualisation, on top of an already high investment” and “Cloud is nothing more than a further step … a continuation of trends”.

So lets look at this “cloud” stuff. It’s pretty hard not to read about it these days … virtual flavour of the month as far as I can tell. However I have a problem with the name … “Cloud” is just plain too vague to really mean anything useful to the average IT Manager struggling under the pressure of keeping costs down, doing more with less and generally getting stressed at his/her lot in life.

Try going to your boss and telling him/her … “we need a cloud in the backoffice to enable virtualisation of all server and desktops to provide software as a service throughout the corporate infrastructure”. You’re likely to get a “clout” across the ear rather than a “cloud” as you intended.

The term “cloud” is great for journalists, bloggers and politicians … talk all day without saying anything, but it doesn’t seem to have jelled too well with Joe Public Computer Professional and it certainly isn’t finding a home with the economic decision maker in business … your Finance Manager for example.

So to lessen the confusion, clear the indecision and illuminate all those in the dark, I’m suggesting a new term be used instead of “cloud” … “large vague grey computing concept”.

That way, the magazine headline could have read: “Gartner: Aussies to shun large vague grey computing concept this year”.

Now that makes much more sense.

Ciao
Neil

Kash posted the following as a reply to a Tom Treadway post. Since out of respect I won’t continue Tom’s posts I’ve made a new post in regard to this …

I took 7 SATA 300 drives of varying sizes (2 x 640GB, 4 x 500GB, 1 x 1TB) from different manufactures (WD 640, Samsung 640, Maxtor 500, 3 x Seagate 500, and I can?t remember the 1TB). In a Linux 2.6.29 machine with 2 x Intel QX9775s @ 3.0GHz, 8GB of RAM (4 x 2GB @ 1066) and an Intel DX5400 motherboard, I set up an LVM stripe over all of these devices. It didn?t allow the full size to be used (it defaulted to 7 x 500GB drives), but when I benchmarked the array, it saturated the PCI-E 2.0 bus. Yes. The drives pushed out 500MB/s constantly. I?m sure if I had PCI-E 3.0 (not out yet) it would have seen even higher speeds, considering that each drive (save for the Maxtor 500) pushes about 110-120MB/s on its own.

Kash … I need a lot more information before going to much further with this one. Firstly, need to know what card you are using. Since the PCIe specification doesn’t have a 500MB/sec limit - it jumps straight from 256 for x1 to 1GB/sec for x4, I’m not sure I see where your limitation is.

Since most of our cards are PCIe x8 they have a realistic throughput of 1800MB/sec (give or take a few mb). What sort of benchmark you are using, and whether it’s streaming reads or writes or random reads or writes will have a major impact on performance.

We have customers attaining well over 1000Mb/sec streaming read speeds from SATA drives these days, so there’s plenty of performance left for you to go yet.

Ciao
Neil

It is with much sadness that I note today the passing of Tom Treadway. Tom was one of the founding fathers of this blog, and a giant amongst his peers in the RAID and storage industry.

Tom was one of the two founders of DPT and a director and principal engineer at Adaptec for many years.
Many of you have gained greatly from Tom?s insightful and casual manner in simplifying complex RAID questions down to simple answers that all could understand. His genius lay not only in his subject knowledge but also in his easy-going manner and simple way of explaining things. I have never tried to continue any of Tom?s threads in this blog ? that would find me standing in boots way too big for a simple soul like myself.

Since Tom left Adaptec several years ago he has struggled valiantly against a debilitating disease to which he has now succumbed.

Please join me in remembering Tom. If you have comments you wish to forward to his family please blog them here and I?ll forward to the appropriate channels.

This is truly a sad day for all.

Regards,
Neil

While it’s good to have good backups, most system admins would like to know just exactly what the consequences of component failure mean for their uptime/downtime/workload scenario. In other words, how big a job will it be to get the server back on line if something falls over.

So let’s look at how the storage subsystem should behave in the event of catastrophic component failure.

The easiest thing on the list are the hard drives. If they fail you should hear a screaming noise from your raid card, you should get an email from your management software (Adaptec Storage Manager) and you should see a red light on the failed drive in the case of a hot-swap chassis. Pretty simple really. Just replace the drive. The RAID card will (depending on settings in the BIOS) automatically rebuild the array, or sit there until you make the new drive a hot spare, in which case the card will then rebuild the array.

We’ll come back to the subject of drive failures towards the end of this because there are issues with drive failures and different RAID types.

Backplane failure? Pretty rare but it should just be a case of replacing the backplane, reconnecting everything and all will be good. The real problem here is that the RAID card is now more than a little annoyed at have had all it’s drives removed and will have to work out what was on the drives. This is simple if the drives all dropped off the card at once. However if they went down in sequence then the card will, at some point, have marked the array as failed and you’ll need to talk to tech support about your options (too long to list here but you are not totally without hope).

Card failure? Just replace the card. Adaptec store their RAID data on all the drives plus on the card itself. The first thing the new card will do is read the metadata from the drives, load the array information into the NVRAM on the card and you’re away. Note that you may be prompted to accept the finding of a new array, which confuses people. It’s not new to you, but it is new to the new card.

Motherboard or total system failure. Just replace the components and your storage will be fine. You can even take the RAID card and array (disks) to another system, plug them in and the card will still know the array and present the data … it’s up to you to work out any OS issues that this causes but it’s generally not the end of the world.

Going back to RAID failures … of course there are diffent RAID levels which have different redundancy capabilities. If you are nuts enough to use RAID 0 you have no insurance - one drive failure will kill everything - don’t bother ringing us, we can’t do anything for you.

RAID 1 can survive 1 drive failure (after all, there are only 2 drives in there). RAID 5 can survive one drive failure - the real problem here is that you can sometimes be caught with a second drive failure during the rebuild after the first drive failure. If the rebuild is not complete then this is regarded as a two drive failure, and you’ve had it. (Note I’ll do a separate article about the dangers of building too large a RAID 5 array later.)

RAID 6 can survive two simultaneous drive failures, so it’s safer than RAID 5 because if you have one drive fail, replace the drive and start the rebuild and another drive fails before the rebuild is complete the system will survive. You’ll be very annoyed, but you’ll still have your data. RAID 10, 50 and 60 can survive varying numbers of drive failures, but you have to be lucky which drives fail. In general you don’t want to count on being able to survive multiple drive failures in these configs, but most of the time you can.

Of course we recommend you have a hot spare in your system. This is just a drive sitting there watching all other drives (but doing nothing itself). When one drive dies, the card will initiate a rebuild onto the hot spare minimising your downtime. While the system is rebuilding you can trot off to the shop to get the failed drive replaced etc. When you get the new drive you replace the failed drive then make the new drive the hot spare. Don’t forget to move the nice neat printed label you put on the front of the system indicating which drive is the hot spare. Do not, under any circumstances, try to re-arrange the drives so that the hot spare is back where you originally had it, either physically or per drive ID. We can handle a bit of randomness - it’s humans that just have to be neat.

So as you can see, you can survive a fair amount of damage happening on your system without your world falling to pieces, but that does not, ever, mean you don’t need good backups. My mate Murphy was an optimist … if you have good backups then nothing much ever goes wrong. It seems to me that if you don’t have backups then fate kicks you at the worst possible time.

Ciao
Neil

There seem to be two types of speed-freaks in the storage world … streaming (video) and database afficionados. I thought I’d point out a few issues that the database lads should think about when setting up their systems.

There are a couple of issues where database differs from streaming … firstly in drive type. If you are running SQL then you need SAS. I once tried to get our marketing boys to go with “SQL=SAS” in an advertising campaign, but they just didn’t seem to think it had enough sex appeal. Whatever the appeal level, it’s a pretty basic but important fact … if you are running a database you need fast drives with low seek times - therefore 15K SAS. SATA drives are great for basic fileserver and video streaming, but they don’t cut the mustard in database.

The next area of consideration is RAID type. Don’t use parity RAID for database. By that I mean that RAID 5, 6, 50 and 60 are out for database … they are fine on reads, but due to the characteristics of these RAID types their speed on small writes (which database tend to do a lot of) are slow.

Instead, you need RAID 10. Some people will use a mirror (RAID 1) which is OK at a pinch, but RAID 10 is best. It is very fast for both reads and writes, gives good redundancy and is scalable in units of 2 (start with 4 drives and scale up by pairs).

The last piece of the puzzle is something you may not be aware of. Adaptec have an “OLTP” performance option on their Series 2 and Series 5 Unified Serial RAID cards. OLTP stands for on-line-transaction-processing, which is the marketing lads way of saying “database” (since they have trouble speaking real English).

This setting changes the underlying caching characteristics of the card to optimise it for the types of reads and writes that we see from database applications. Cache page size, write cache, IO sorting and prefetch (read ahead) are all modified. These settings are not user-configurable, it’s just a set and forget option.

We’ve seen performance gains up to 30% in some environments which is a performance gain not to be sneezed at. There are further advances in technology coming soon to improve database speed even further but I’m not allowed to talk about those yet

As you know I’m an eternal optimist and only speak of upsides, so what are the problems with going down this path? Firstly there’s the mixed environment server. If it’s doing streaming and database then this probably won’t help. As a matter of fact if it’s doing both these tasks then nothing much will help. The other “downside” that people come back at me with is the cost of RAID 10, especially on SAS drives.

Sure, it uses a lot of drive space and costs heaps, but you have to compare that to saving money on building the server vs having your users wait a second or two every time they access the database. I’ll leave it up to the accounts to do the year-on-year calculations but I know that people ringing me up for advice want speed, and this is the basic formula to get just that out of your database.

Ciao
Neil

(I’ve been away from the keyboard for a while, and needed to get the fingers working again, so I’ve stayed away from anything new we are doing so as not to upset the marketing lads any more than usual and have gone for something simple to get the hang of this again )

There are two ways of adding capacity … I call them horizontally and vertically. By horizontally I mean across the array … adding more drives to an array to increase it’s size.

Adding capacity vertically is what I call the ability of a RAID controller to increase the size of an array by using additional space that may exist on the disks already in the array.

Thinking horizontally …

The most common expansion request we get is whether you can just add a drive to an array and increase it’s size. The basic answer (without all the caveats) is yes. Add the drive to the system (hot swap is always good as you don’t have to bring the system down), then go into Adaptec Storage Manager, right-click on the array and “reconfigure the array”. Simple. Choose all the drives, RAID type and set your capacity.

Thinking vertically …

This one is not so obvious. It’s basically the same approach - reconfiguring the array but just choosing a new size. Lots of people just seem to miss the fact that you can use the same disks etc, just increase the size. Of course there needs to be space on the disks, but you might have deleted another array freeing up space, or, heaven forbid, changed the drives over to larger disks by using the dreaded “replace one at a time and let the array rebuild” method. Note that there is also another way that we don’t advertise to migrate an array from one set of drives to another set, but I’ll put that one in a separate blog.

Either way, if there is space available on the disks you can grow the array without any great hassle.

Note that you can do this live without shutting down the system (read the caveat about hot swap drives). The RAID card will just chug away in the background expanding the array while you are still using the system. Will there be performance degradation? Yes. Will it kill your system? No. Do it over the weekend.

Now the caveats …

HostRAID cards don’t go past 2TB. If you want >2TB from Adaptec look at our 2, 3 and 5 series Unified Serial cards.

There are a couple of major issues here which revolve around the OS, not the RAID array. XP 32 does not support greater than 2TB - end of story. Other, newer Windows OS do support greater than 2TB (check the limits with the vendor), but not for the boot volume. There is nothing you can do about these limitations, and for that matter there is nothing we can do about it either.

2TB? Sounds big, but today it’s an absolute no-brainer to break this limit so check out your OS as well as your RAID card before going big.

There, 10 minutes of typing and I feel all warm and bloggered again - that’s better.

Ciao
Neil

Expensive RAID controllers used to be the domain of business. You just didn’t find powerful, expensive controllers and lots of noisy, hot, large disks in home machines.

However things seem to be changing. People are putting high-end RAID controllers into home machines, combining them with SSDs to produce massively fast machines which do not run incredibly hot or make lots of noise, making them suitable for home machines. They seem to be choosing the high-end controllers to match them up with the perceived high performance of their SSDs.

This poses a problem. All manufacturer’s RAID cards have traditionally been put in server system. They then rely on high-volume airflows within those machines to run air across the card’s heatsink to cool the card. Servers, especially rack-mount units, generally don’t have any problems with high air-flow volumes … but they have plenty of problems with noisy fans. You don’t generally find 2U rackmount servers in the family home.

With home machines the noise is a problem. There is traditionally not significant airflow at the card location within those chassis. If there are fans, they tend be aimed at the hard drives, not the expansion cards. So what are you doing to keep things cool? Do you have sufficient airflow across your card? What is the temperature of your card?

I’d be interested to know what’s happening out there.

Which of course means I don’t have a life, but …

I once wrote an article called “using high-port-count controllers”. About the only thing I was unsure about was where to put the hyphens. It was all about mixing SAS and SATA hard drives on high-port count controllers. The usual semi-sales pitch for our 52445 product etc. The mentality was basically … use SATA drives for slow stuff, SAS drives for fast stuff and because of these different RAID arrays you’ll need lots of ports.

Simple really.

Then along came Solid State Drives (SSD).

My initial reaction was to go back to Word, use Control+H to replace SAS with SSD and republish (which is what a few of my colleagues seem to be doing). But it turns out there’s more to think about that just replacing SAS drives with SSD drives.

I’ve been talking to more and more people lately who are playing with SSDs … especially the Intel type (flavour of the month so to speak). Surprisingly I’m finding these drives not in the big servers running the massive databai, but in much smaller, often workstation, performance machines.

Now my initial reaction to people questioning me about putting 4 SSD drives on one of our controllers was to ask them what they were now doing with their now unused SAS drives (I was looking for a few donations for a fast system for my kids). Surprisingly the now almost standard answer has been … no, I was running on a single (or couple of) SATA drives before this, often on on-board software RAID.

Wow. I thought moving from expensive SAS drives to obscenely expensive SSDs was a big step, but going from a single-disk workstation to 4 SSDs on a RAID card is amazing - chalk and cheese. Therefore I’ve been asking quite a lot of questions of people, mostly based around the word: “Why?”

Speed (speed, speed and more speed)
Reliability
Speed
Low running costs
Speed
Cool running systems
Speed
Quiet systems
Did I mention speed?

Now you could build a fast system before SSD drives came along. 15K SAS drives have been around for a while, and you could put a pretty quick system together if you have the right skillset and enough money. But not that many people did it. The world tried to use SATA drives to everything. SAS remained relegated to the high-speed database servers as these were the only machines it seemed that warranted such prohibitive expense.

However, SSDs seem to get a different reception from the general public to the old SAS drive. “Yes they are expensive, but they’re fast!” Well so are SAS drives, but Joe public didn’t put them in his workstation.

It seems that Joe public views the SSD drive like the iPhone … not really sure I need one but it’s new, funky, cool, green, the neighbor has one so therefore I have to have one. I’d love to hear from people as to what use they are making of SSD drives. Are you using them because you have a database with 14000 users which just cant run fast enough on SAS, or is because you just want to be at the bleeding edge of drive technology in 2009?

Drop me a line and let me know what you’re up to. It will be interesting to see what different ways people are using this technology.

Ciao
Neil

No, this is not a relationship or lonely-hearts column ? it?s about technical specifications for products ? and when you can?t find answers to basic questions like ?what processor does it have onboard?.

Adaptec have for years spent their life quoting statistics and specifications for their products. We have people who speak a strange sort of language that is unintelligible to the normal human being ? 1.2Ghz processor with 512mb of RAM capable of x number of IOPS and so many Mb/s throughput ? so on and so forth etc etc etc.

So when I receive an email from a customer asking me to compare product x with one of our cards (call it product y), and for me to tell them why product x is so much cheaper than our card. Now I don?t normally worry too much about price, because (a) that?s the salesman?s job and (b) when you compare apples to apples we come out OK.

Now this is a pretty normal occurrence for me ? quick flick of the competitor?s website, read the specs, send back some salient points to the person who asked the question, then move on to the next email without putting too much thought into the matter.

So I check out the competitor?s website on their card. Looking at the capabilities this is a pretty good product (RAID 0, 1, 10, 5 and 6). No mention of RAID 50 or 60, but that?s OK (gives me an advantage). Lots of memory, and options for more with a battery. Hmmm, sounds good so far. Let?s check out the processor to see what sort of grunt this thing has to do both day-to-day parity calculations and, just as importantly, rebuild arrays after drive failures.

Check the main page, check the tech specs, read the fine-print, look at web-based reviews from third-parties ? no mention of processor speed. Now that is, to put it bluntly, strange. If this competitors marketing team is anything like mine, they?d never let a page go by without putting something as basic as a few Mhz on there. But ? no, not a mention anywhere of the processor speed.

So where does that leave me? How do I answer the customer?s query? Simple really ? the competitor doesn?t mention the processor because it?s not there. It?s a ?hostraid? card ? not a hardware RAID card. Nothing unusual with this, except it?s being promoted as a competitor against our hardware-processor based cards.

Is this unusual? Not really, but it brings to light that if today?s heavily-pressured marketing departments don?t say something it?s generally not a mistake ? what they don?t say counts as much as what they do say.

So make sure you are comparing apples against apples next time you check out two products.

I spend a bit of my time showing Adaptec products to people throughout different parts of the world. Our current push is on our iSCSI/IPSAN software-only product called OnTarget. Now that sounds like a plug for a product (which of course I would not do here), but it sets the scene … something new, something you may not have heard of before and something you may not be 100% familiar with.

This is really about looking at how people perceive products, use products, and react to the way other people use products (when it’s different to the way we perceive how a product should be used).

I see people use our high-port-count SAS/SATA cards by direct attaching drives … I mean 24 SATA drives direct-attached to a RAID card … and I cringe. I don’t say anything but I think to myself … what drug is this guy taking? Of course, the usage is perfectly acceptable, and in fact is a cheap way of making massive storage volumes, but it just sits a little outside my comfort zone … I think those disks should be in a JBOD, but that’s my personal opinion and personal experience tainting my view of how something is done.

Just because I think it’s nuts, does that mean it’s wrong? No, but I reserve my right to think it’s nuts. I just shouldn’t say it to anyone.

So how does this fit with OnTarget? Simply that I find many people have a preconceived, set view of where iSCSI/IPSAN fits in what section of the storage market. In other words … it has a place and that is where it should stay. That doesn’t just mean customers. That also goes for industry experts. I constantly read that iSCSI/IPSAN is for entry-level and you need to move to fibre for enterprise.

What a load of (insert your own expletive here). People put iSCSI/IPSAN in the entry-level sphere because that’s where they were told it fits. They don’t bother thinking about novel, or different uses of the product, or the fact that it’s evolving a truck-load faster than it’s competitors, they just pigeonhole the technology straight into ‘entry-level’.

Welcome 10gb ethernet. Does it run at 10gb throughput? No. Is it dramatically cheap right now? No. But put two of these ports together into the storage device and is it fast enough for enterprise? You bet. Is it cheaper than fibre and getting cheaper by the day? Dead right as well.

So if you think a little about iSCSI/IPSAN, and where it fits, you need to take into account it’s changing face. 10gb ethernet will give somewhere around 6-7gb throughput (600-700mb per second). Put that across 2 ports in a teamed connection, and now you have some real performance on your hands. Combine that with dual-mirrored systems and now you have a complete fail-over system running at speeds only previously dreamed of in the expensive fibre world.

Does that sound like ‘entry-level’? It doesn’t sound anything like it to me, but next time you read an industry analyst’s positioning of iSCSI/IPSAN, look for ‘entry-level’ … some people just refuse to look a second time.

See … if you put a single system together with SATA drives, a small amount of system RAM and gigabit ports, then yes, it’s an entry-level system. However if you change that same system to SAS drives, 4gb of system RAM and 2 x 10gb ports then it’s a ball-tearer that fits perfectly in the enterprise space, especially when mirrored into a dual fail-over system. All, of course, at a price that will make the fibre lads weak at the knees.

Case of weirdness in point … I’ve seen people create a Windows RAID array on DAS drives, an iSCSI volume on an external device, then mirror those volumes in the Windows OS. Now I thought that bizarre (and it probably was), but the user’s logic was to be able to take that volume to any system in the network and mount the volume in the case of a disaster. Several years ago I thought that strange … today I’d expect that to be a standard function of an iSCSI/IPSAN storage device. So what was out of my comfort zone has become a norm.

Here endeth my lesson. Take a long hard look at the way people use equipment. It may not be exactly what you need right now, but it may well point you in the right direction. Just because someone (like me) says it’s wrong, look twice and make your own decision. Likewise, when someone pigeonholes a technology into a certain market segment, look twice, and ask yourself whether in fact that is true, or just another editorial hack promulgating the same old same old.

iSCSI/IPSAN fits both entry-level and a fair old chunk of the enterprise … you just have to look past the BS coming from industry experts and take a second look at something you previously believed below your requirements.

Food for thought.

I’ve just read an article in one of my favourite (and very good) trade mags … all about going green … the so-called “GoGreen Journey” (another swanky name invented).

With such interesting paragraph headings as “leading the way”, “carbon footprint analysis”, “training employees”, “planned marketing” and “inspire” made me think that maybe this one was worth reading.

However, as usual, I was sadly disappointed. The usual green edtorial waffle that fills page after page and says nothing.

Whatever happened to “Turn the lights out when you leave!”, “Turn your PC off before you go home!”, “When you purchase a device make sure it has some form of energy saving capability!”.

Where has all the commonsense gone … being green and saving energy/money is simple … JUST DO SOMETHING … stop wasting paper waffling about nothing.

Between the “Global economic crisis” (don’t get me started on that one) and the “GoGreen Journey” (new name for common sense) I’m almost driven to reading something interesting like Backup Software Manuals … almost.

Windows Small Business Server is an extremely popular complete operating system for small to medium business throughout the world. It contains most applications that a business requires in a simple to install and manage package.

This article concentrates on the storage needs of SBS, in particular creating the correct RAID arrays to get the best performance and flexibility from your system. It goes without saying that you require a system with sufficient CPU and memory capacity, but that information can be provided by any system integrator.

Horses for Courses ? different raid types for different types of data
While there are many different ?types? of RAID arrays, they fall into two basic categories ? Parity and Non-Parity RAID arrays. Each has their benefits and detractions, with each type being suited to different kinds of data.

Non-parity RAID
Non-parity RAID can be simply defined as RAID 1 (mirror), RAID 10 (stripe of mirrors) and RAID 1E (stripe of mirrors on odd numbers of disks).
The benefits of non-parity RAID is that they have very good write speeds, especially for small data writes. This makes them very well suited to support Operating Systems and any kind of database files.
The downside to non-parity RAID is the cost ? they take up considerable disk space. For example a RAID 1 (mirror) gives 50% usable space from the disk drives (2 drives gives 1 drive space).

Parity RAID
Parity RAID can be simply defined as RAID 5 (distributed parity capable of surviving single disk failures), RAID 6 (distributed parity capable of surviving 2 disk failures) and RAID 5EE ? a different kind of RAID 5 that contains both parity data and hot spare disk space.
Parity RAID is the great all-rounder of RAID arrays. RAID 5 is generally good for most types of read and writes, and gives quite good streaming write speeds, but it suffers from performance problems when writing small amounts of data (such as database and operating system writes). It is the most economical of RAID arrays in that the equation for capacity is N-1 (you lose one disk?s worth of capacity). For example, 3 drives in a RAID 5 will give 2 drives worth of capacity. 10 drives in a RAID 5 will give 9 drives worth of capacity.

Hot Spare Disks
A hot spare disk is a hard drive that does not contain data, but is simply spare to the system. When a drive fails a hot spare disk will ?kick-in? and replace the failed drive, rebuilding all the RAID arrays affected by the disk failure in as short a time as possible.
This is very important. It minimises the time in which the system is at risk of a second drive failure (which in most cases would cause data loss). The system rebuilds automatically putting the RAID arrays back into a safe state, allowing the administrator time to source and replace the failed drive.
So if there is room in the system, and the customer can afford the implementation, then put a hot spare in the box. Especially if the customer site is well-away from an easy supply of hard drives (remote location), it makes a great deal of sense to have the hot spare kick-in and rebuild the array as quickly as possible, giving the system administrator time to source that replacement drive and re-implement a hot spare.

Different types of hard drives
The two basic types of hard drives available on the market today are SATA and SAS.
SATA drives are cheap, fairly reliable and have a large capacity. They are commonly found in desktop and laptop systems, and are now finding their way into servers because of price and capacity requirements. The thing to note about SATA drives is their inability to do many things at once. They are single-tasking devices with fairly slow response times ? in other words they are great for streaming but in database applications they suffer from not being able to react at the speed of their more expensive cousins.
SAS drives on the other hand are smaller, much more expensive and very much faster. SAS drives are excellent for database and operating system work because they do multiple things at once (multi-tasking) and have very fast response times. In other words they react very quickly to the small reads and write requests from databases.

Different types of drive controllers
Inside all servers is (or should be) a RAID controller. A RAID controller gives a system the ability to survive drive failures without losing data, and allows a system administrator to create different performance sections within the one system.
RAID is a compromise between performance, reliability, cost and capacity (pick any 3). While I say it is a ?compromise? I don?t mean this as a detraction. RAID controllers are essential to server systems to ensure continuity of data to an organisation (and help minimise expensive system downtime).
Some manufacturers create RAID controllers that connect only SATA hard drives (SATA controllers). These are limited in the number of drives they can attach to and will only work with SATA drives.
Adaptec RAID controllers are ?SAS? controllers. This means they talk to SAS drives AND SATA drives. In fact you can connect both SAS and SATA hard disks to the same RAID controller. This gives great flexibility in building a system by using drive types that suit your data needs, but we?ll talk about that later.

The components of SBS
These can be fairly simply defined as:
The Operating System
Exchange
SQL
Data
Of course there may be other specialised applications a user runs on Windows SBS, but for the moment we?ll concentrate on the components that come with the standard install package.
Why is this an issue? The different components of SBS have different data characteristics. You can basically put them in two categories ? database-type data and general-purpose or ?streaming? data.
With SBS the operating system (Windows Server), Exchange and SQL all have database-type characteristics ? many small read and writes to the underlying disks in a completely random manner.
The data portion of the disks (Word, Excel, Powerpoint, customer data) are generally less dependent on speed, with large capacity being much more of an issue to the system administrator and end-user.
Basically, with SBS, what you want is a disk structure that is good for database, AND a disk structure that is good for general storage ? however they are not the same thing.
Building RAID arrays ? the old and new ways of the world
In the past, and still with many of our competitors, the building block of a RAID array was a complete disk drive. This meant that if you wanted to create a mirror of 2 x 250gb drives, the mirror consumed both drives completely and no space was left on the drives for any other use.
Similarly, a RAID 5 on 4 disks used the complete space of all 4 disks ? giving no flexibility to the system builder do anything else with those drives.
Now if you had an unlimited number of hard drives to play with then this would not seem much of a problem ? however that is not the economic or practical reality of building computer systems today.

Adaptec RAID controllers use a different methodology to building RAID arrays. Instead of the basic building block of the array being a complete drive, it is now a portion of the disk (being whatever size portion you want to use). This is commonly called a ?container?.
Therefore a RAID array on an Adaptec RAID controller can be made up from small containers on different hard drives. It is possible to create different RAID types on the same set of disks using this approach.
For example ? if I had 4 hard drives in my 1U server, I could create a RAID 10 of say 50Gb capacity across all 4 hard drives. This would in fact use only 25Gb from each of the hard drives. The rest of the space on the drives can then be utilised by the RAID card to create other RAID arrays (of the same or different types).

Putting this all together ? the problems
SBS is an unusual software package ? it places demands on a system for both database and streaming-type data, generally on lower-end server system. By its very nature Microsoft have created a system that users see as being a cheaper solution than purchasing all the individual components, with many users carrying that cost saving mindset across to the hardware that the system is installed on.

This pretty much means SATA disks. While SAS disks would be much better, they are generally too expensive for the type of customer who is installing SBS, especially if the customer is looking for capacity as well as performance.
While I have nothing against SATA disks, and they are perfect for streaming and general-purpose data, they are not ideally suited to database applications. Now as we have read before, SBS has some applications that have database-type read and write characteristics, and some applications that are much more general in nature.

This means that running SATA drives puts the database components at a disadvantage. To counter this it is essential to run these components on a non-parity RAID array (preferably a RAID 1E or RAID 10).
However, RAID 1E or RAID 10 consume a great deal of disk space (50%), which makes them much less suitable for the general data storage that comes with the majority of the customer?s data.

So here is the problem. What we really want is SAS drives running non-parity RAID for database applications, and SATA drives running parity RAID for general data storage. All of a sudden to get the best performance and flexibility from your SBS system you need to spend a lot of money on the server hardware. This flies in the face of saving money by using SBS in the first place.

The solution (or at least one of them)
The majority of users will use SATA drives for their SBS system ? price and capacity dictate that this is the most sensible way to go.

So ?
Using an example of a server with 4 x 500gb hard drives, the following scenario would be an ideal implementation for an SBS system.
1 x 50Gb RAID 10 array for the Operating System
1 x 100Gb RAID 10 array for Exchange and SQL
1 x 1.2Tb RAID 5 for general data

The benefits of this implementation are:
Both the OS and Exchange/SQL applications reside on a non-parity RAID array (RAID 10). This gives them the best disk performance that can be achieved with 4 disks (while still retaining data safety).
The data volume would reside in the large RAID 5 array. This gives good performance for general file serving, while maximising the capacity of the remaining space on the drives.
Using an implementation like this, a user can get the best of both worlds. High-speed, non-parity RAID arrays for the OS and database applications, and a large parity array for their general fileserving data.
It is important to note that this feature resides on all Adaptec 3 and 5 series RAID controllers. Competitors use a different technique ? called slicing, to break a raid array up into different logical volumes (drives) for use by different parts of the operating system. The problem with this approach is simply the fact that the underlying RAID array is the same for all logical drives. So if a user chooses RAID 5 as the underlying RAID structure, then all disks are running on RAID 5 ? which as we have read is not good for database. Adaptec have taken the approach that the RAID structure should be flexible, and a user should be able to match the RAID type to the data structure sitting on those disks.

The problems of growth ?
Building a server that will have sufficient storage for a customer for the entire life of the server puts a large financial burden on the initial build of the server ? paying for a lot of disks that you aren?t currently using is not a good use of business cash reserves.
So ? start with the capacity you require. You can add drives to a system at any time, and expand the RAID arrays on your existing drives to incorporate the new drives. There are, of course, different caveats on this with different RAID types, but RAID 10, RAID 1E, RAID 5 and RAID 6 especially are idea candidates for adding drives and increasing the size of the system.
Does this take a long time? ? possibly (depends on the size of the disks). Does this mean a lot of downtime? ? No. Once the disks are added to the system then it can be restarted and this expansion process completed in the background, while users are working on the system.

But what if I only have 3 disks?
3 disks brings about a slight change of plans to the solution described above. I?m no fan of mirrors ? they are safe but not particularly fast, and they only use 2 of the 3 drives available.
In this case instead of the RAID 10 described in my first solution, you would use a new type of array called 1E. It?s basically a mirror on 3 or more drives. A mirror on an odd number of drives? Sounds weird doesn?t it, but it works, and is faster than a mirror because you have more disks working for you rather than the 2 of the mirror. This also works in my above solution because 3 disks will handle the RAID 5 capabilities of general file serving for SBS.
Using different disks types for ultimate performance/flexibility

My ideal solution for SBS (or just about any other server system for that matter), is to both combine RAID types and disk types in the one system. SAS drives are very fast, with rapid response times in database and OS applications. Therefore, my idea solution would be 3 drives with two RAID 1E created on them ? one for the OS and another for the database portion of SBS.

I?d then add 3 or more SATA drives in a RAID 5 or 6 to create a large space for general fileserving duties. This combination places the database-type data on fast responding SAS disks, but keeps the cost down for the large capacity file-serving duties of the server.
Card types this works with
Adaptec?s Series 2, 3 and 5 cards can all create multiple arrays on the same set of disks. Note that the Series 2 is an entry level hardware RAID card that can?t do RAID 5 or 6, but the Series 3 and 5 can create just about any RAID type you wish on as many drives as you wish to attach.

What happens when a disk dies in this scenario
For the technically-minded folks who?ve made it this far through this blurb, you?ll be wondering ? what happens if a drive dies when there are 2 or more arrays on those disks. Nothing unusual or wonderous happens at all. All arrays that are located on the failed drive are impacted but keep running.
When the drive is replaced, or when the hot spare kicks in, all arrays are rebuilt and life goes on as normal.

Summary
(It?s always nice to see that word ? means the ramble is almost finished) ?
Windows Small Business Server and Adaptec RAID controllers make ideal partners. Adaptec?s ability to create different RAID types on the same set of disks helps get the most performance and capacity to enable SBS to fulfil the customer?s needs.
If you learnt RAID 5 years ago then it?s time to go back to school. A lot has changed. Multiple arrays on the same disks ? different disk types in the same system ? expanding systems on the fly while customers are still working ?
Maybe it?s time to talk to your Adaptec RAID team and get the full story.

I talk to a lot of people who know a lot of stuff about servers. They sell them, build them, repair them, attach storage to them etc. So when we start talking about external storage in general and IPSAN in particular, why is it I get the feeling there is such a vacuum of specific knowledge out there?

ISCSI, IPSAN … it’s a foreign language to a lot of the IT industry … why?

I’ll spend one meeting talking to people who know more about the ISCSI protocol than I regard as healthy for the average human, then spend another talking to people who are brilliant at recommending server solutions to customers using big brand name equipment.

Why then, should it be, that the people in one meeting have absolutely no idea what the people in the other meeting are talking about?

Don’t get me wrong … I’m not saying that sales people should have a brilliant handle on the technicalities of their products (they are, after all, sales people), and I don’t think that the technical boffins in the backroom should be able to sell jack, but in every other sphere of computing there is a middle layer who understand the technology and can position it within a marketplace.

So why can’t we do it with IPSAN?

Application engineers are a dime a dozen in serverland, networking, disk storage, even to some degree fibre san, but they are not exactly easy to find in the IPSAN arena.

This technology has been around a long time. It’s mature, it works and is ready for the general masses. So what is stopping it. If I read another analyst report saying that IPSAN will grow by x next year (and will be the next big thing) I think I’ll puke … it’s been that way for years now and never seems to change. Why?

I believe it’s because we are lacking people who (a) understand the technology to a reasonable degree, (b) can explain it in laymans terms to the end user and (c) can dispel the fear and doubt that the fibre lads have been dispensing since the usurper came along and started to challenge their right to make obscene profits.

So here’s the challenge. Send me your storage scenarios. Let’s make some discussion about the pros and cons of IPSAN and see if we can find people out there who can explain its benefits and weaknesses to the general public in a language they can understand.

I know there is a gap in the IT knowledgebase in the area of IPSAN and ISCSI … let’s fill it in a bit.

Otherwise we’ll be reading those same reports again next year.

(I think I?ve watched the Karate Kid too many times)

System builders are often faced with the dilemma of adding enough drives to a system to meet a customer?s capacity and performance requirements. The traditional approach has been to purchase a chassis large enough to hold all the drives internally. This approach has a few problems.

Firstly, you need to spend up big on the initial chassis. Even if the customer doesn?t need all that space right now, if you are going internal then you still have to spend enough money up front to purchase the 24 or 48-drive chassis.

A smart customer will want to purchase a system that can grow with his needs, but purchasing a massive server up front often puts the quote out of the reach of the initial sale.

Purchasing a smaller chassis without expansion capability often means people are thinking of replacing the hard drives at a later date with larger drives to meet a capacity upgrade requirement. Bad move ? this is not cost effective, takes a lot of time and is not in the best interests of the customer (even though it may get the system builder a bit of extra income at some future point in time).

So ? you want to build a system now that meet?s today?s requirements for capacity, and you only want to pay for what you are using now ? smart. But ? you know your space requirements will grow in the future and you don?t want to throw away a perfectly good server just because it has run out of disk space a year after purchase.

What to do?

It?s pretty simple really ? just put the disks ?outside? the server. Many users can purchase all the horsepower and memory requirements they need in a cheap 1U server. By adding a JBOD (just a bunch of disks ? I love that acronym) and a RAID card with external connectivity, you can add as many drives as you like. On top of that you can just keep daisy-chaining JBODs together until the cows come home, giving you as much storage capacity as you can possibly imagine (try 250TB in the one server).

If this is so simple, why do people shy away from it? FUD (fear, uncertainty and doubt) is generally the reason. Let?s look at some of the questions people have when considering this type of configuration:

? I?ll create a performance bottleneck between the server and the storage ?
? If one cable breaks I?ll lose all my storage ?

A modern SAS RAID card (and yes, it accepts both SAS drives and SATA drives at the same time), has a funny looking new type of external connector called a 4x (4 by, 4 lane, 4 channel ? call it what you will) connector. This is, practically, 4 single SAS/SATA communication channels joined together at the hip.

A single SAS/SATA connector runs at 3 gigabit ? roughly 300mb per second throughput. But SAS is smart, much smarter than SATA. When a SAS card finds a 4x cable going from the card to another device that also uses a 4x connector, it joins all 4 cables into one large pipe. Sort of multiplexing for those of us old enough to remember that technology.

The end result of all these smarts, is that the pipe between the RAID card and the JBOD is capable of sustaining 1200mb per second throughput. That?s pretty fast, and eliminates any bottleneck concerns that a user may have about connecting drives externally to a card.

Now the question about the cable. When was the last time you saw an external cable ?break?. I?ve spent a lot of years in tech support, especially dealing with SCSI. Now SCSI connectors were dodgy to say the least, and were (and still are) often used on devices like external tape drives where they were connected and disconnected frequently.

Over the years I have hardly replaced a SCSI cable. Now if we add to that the fact that modern SAS external cabling uses a much better connector type and cables connecting servers and JBODs are not connected/disconnected frequently, and you end up with an almost bulletproof connection system that is by no means the weakest link in the data storage chain.

Now that we?ve dispelled the myths (FUD), let?s look at the benefits of building a server system in this manner.

Firstly, you start with a 1U server with 4 drives (or 2 or 3 and a CDROM). Spend your money on CPU and memory components. In that box you need a SAS RAID card that has 4 internal connectors to handle the (up to) 4 drives inside the box, and a 4x external connector to connect to a JBOD. An example of a card like this would be Adaptec?s 5445.

Then buy yourself a JBOD. These are generally 2U rack units that take 12 drives. You may or may not want to purchase all 12 drives now ? just purchase as many as you need. Therefore your initial investment in the server hardware comes down to a beefy 1U server, a good quality RAID card, a JBOD and as many drives as you currently require.

But time goes on ? and sooner or later you?ll run out of space. Simply purchase more drives and add them to the JBOD. Fill the JBOD? No worries, purchase another JBOD and daisy-chain it to the first one. Start filling that with drives. Run out of space? Just repeat the cycle with another JBOD and keep on adding drives for as long as you wish. The technical maximum at this point in time is 250 hard drives, which for all intents and purposes is unlimited ? I haven?t see anyone come even close to this yet.

So you end up with a big expensive server ? great. However you paid for it over a couple of years, and took advantage of the dealer specials along the way. You also gained from hard drive price decreases. Along the way were hard drive capacity increases (these two go hand in hand like death and taxes).

It?s hard to see anything but benefits to users ? reduced capital expenditure, increased flexibility over a fixed system, access to better technology as it both appears and drops in price.

Sounds like a win win win to me.

This little bloke started the ball rolling on “green”. Since then the environment movement and now the IT game have jumped on the bandwagon.

Trouble is … are we sick of hearing about “green”?

I was doing a product presentation yesterday for Adaptec iSCSI Storage devices, after which I branched into our controller range and our newest feature … power saving.

The general response from the audience was “so Adaptec have jumped on the Green bandwagon as well!”. Is this cynicism justified? Personally I’m as green as the next person … got rid of the gaz-guzzling 6 cylinders for a couple of 4-cylinder econo cars, ride the pushbike where possible, catch public transport when possible (and it arrives), recycle everything possible (haven’t worked out a way to recycle the kids yet but the cheese and kisses is working on it) and changed all the light-bulbs for those environmentally friendly but slightly-dim units that have me thinking I need new glasses.

So I regard myself as “green” (sort of), and should therefore be excited about anything that saves energy. Add to this the fact that the boss pays me to go out and promote Adaptec’s new “green” philosophy and you end up with a “green” evangelist running around telling everyone they should “save power” wherever possible.

Trouble is, the general public (as in my presentation audience) have become jaded and somewhat sick of the green message. Therefore I think it’s time to develop a new marketing model for “green”. Try this one …

“Be greedy, save money!” Morally it’s not all that responsible, but logically it makes sense. If you can purchase two products of similar characteristics, and one will save you money over a period of time, why not be “green”, or “greedy”, and in the process save some money, save the environment and give yourself a slap on the back by proudly stating “our organisation is green!”.

For a long time now we have been looking at the fancy stickers on white-goods and trying to work out which one will keep the beer cold for the least cost for the longest period of time, so why not start thinking about computer components in the same manner.

I note that there are various developments in the power supply arena, promoting efficiency (and you can bet they’ll call them green). What about other computer components? Adaptec have started the ball rolling with their Series 5 and 2 RAID cards by introducing power saving features that will help the card pay for iteself over a period of time, but what about the rest of the components in the box?

When was the last time you asked your hard drive vendor … how much does this unit cost to run? What about the motherboard manufacturers? Or the LCD vendors for that matter.

Adaptec have started the ball rolling in the “green”/”greedy” department for computer components … so where it everyone else up to?

Food for thought.
Kermit’s alter ego

The question of changing the disks in a RAID array often arrives on Tech Support’s desk … and up until recently it’s been a an ugly one to say the least.

The basic scenario is that of a perfectly good working system running on small drives. You might have a RAID 5 array made up of 4 200gb sata drives … everything is working fine except for the fact that 600gb is not enough space any more.

Your options are pretty varied here, ranging from adding new drives to the system and expanding the array, or rebuilding the system completely on new drives, or adding drives in an external chassis connected to the server via a SAS card … OR … (and this one is not recommended), removing a drive to degrade the array, replacing it with a larger drive, rebuilding the array, then repeating the process 3 times to change all drives. This is somewhat akin to crashing your car 4 times to get 4 small dents fixed in the fender … crazy.

The sensible alternative is … move the array to a larger set of drives. To enable this you will need to be able to access all drives off the same RAID card at the same time. This could be done by putting more drives in vacant spaces in the backplane, or by directly connecting drives to the same controller via internal cabling … however you do it both the old and new drives need to be connected to the same card.

Once you can see all drives you can simply modify the array (expanding as you go if you wish), then choose the 4 new drives as the destination rather than the original 4 drives. The RAID card will then move the array from the original 4 drives to the new 4 drives … simple. Yes, it will take some time, but the result is a painless transition to newer, larger hard drives.

As for operating system volume expansion … most OS can increase the size of their volumes to make use of additional drive space (except the system volume), so this should not present many problems. You also need to make sure your operating sytem can talk to the size of the new array … remember XP still has 2TB limits on volumes, but Vista and Windows 2003/2008 are fine to talk to massive drives.

There’s one last trick in this scenario. If you are using an Adaptec 2, 3 or 5 series RAID card (SAS/SATA/Unified Serial) you can install a high port count controller beside your existing 4 or 8 port controller, reboot the system allowing the drivers to load, then shut the system down, physically move the drives to the new larger port count card and reboot … the new card will find the array on the drives, prompt you to import it then reboot perfectly.

Certainly makes life simpler.

… but you can talk about it forever …

Has anyone considered the cost of talking green? I’m being inundated with magazines hyping the benefits of going “green” in IT.

Maybe it’s time people just did it, rather than talking about it. We’re likely to run out of trees for all that paper pulp before we decide the “best” way to go.

There are obvious benefits to greening IT, but maybe the Journo’s of the world should be thinking how to get the green message across in a green manner, rather than generating more waste paper.

Blog on.

What are the benefits/disadvantages of using large hard disks?

With the drive vendors now currently selling 1TB drives (and 1.5TB drives in the near future), what are the advantages/disadvantages of using these massive storage devices?

Large drives are great … you can fit a lot of data into a much smaller storage footprint, but there is a downside. Whether you purchase a 500gb, 750gb or 1TB drive from the same manufacturer, if they are the same technology then they pretty much run at the same speed. That’s great for standardisation as far as the drive vendor goes, but it impacts on user mentality and overall performance when it comes to RAID.

Almost all server builders will use RAID to safeguard against drive failure. The problem is that large drives allow them to build servers that give the customer the space they want, but use fewer, rather than more, drives in the raid array.

It also means that customers are using mirrors (2 drives), or 3-drive RAID5 configurations more often.
The problem here is that the fewer drives you use, the less “workers” you have reading and writing data.
So a combination of small drive counts and slow raid types (mirrors or small RAID5) lead to less than impressive performance in general.
For a really technical explanation of drive counts and their impact on RAID5 see (http://storageadvisors.adaptec.com/2007/07/10/effect-of-drive-count-on-raid-5/)
So are big drives better? Not really … they’re just bigger.

If you are interested in performance as well as capacity, think hard about the RAID configuration you put together, including just how many drives you use to give you the best combination of speed and capacity.

System building is not quite as simple as it used to be!

Does the number of disks in a RAID-5 array affect the performance of the array?
Received a question from a reader wanting to know if the number of Drives in a RAID set affect the performance of the Array. The short answer is yes it most certainly does!
Commonly referred to as the stripe width, which refers to the number of parallel stripes that can be written to or read from simultaneously ...