I was going to call this “size does matter” … but out of respect for the young lady who vets these posts I won’t go there
So this is all about the issue of choosing the right size drives to suit your performance requirements. Now a lot of people just choose the number of drives based on capacity, and what overall size they need, but there are a few considerations that should be looked at when it comes to size of drive and performance. I’ll break this into two sections – spinning media and SSD:
Spinning Media …
As a general statement (bound to cause me some grief), the performance of spinning media is pretty much the same across a family of drives, no matter what the size. If you look at something like Seagate’s Constellation 2 SATA drives, they quote seek, read and write speeds of exactly the same numbers for 250GB up to 2TB drives. So while they get bigger, they don’t get faster.
So why does this matter. Well, in spinning media RAID arrays, generally the number of spindles has a major impact on the performance of the array. For example, a mirror (2 drives of 6TB drives – 6GTB capacity) will not be anywhere near as fast in streaming data as 8 x 1TB drives in a RAID 5 (7TB capacity or thereabouts). The additional spindles in the RAID 5 allow for small reads and writes from each drive, speeding things up considerably.
To do some really rough maths on streaming data – if a drive does 120MB/sec read speed, then theoretically the best speed you can get from the mirror is 240MB/sec and from the RAID5 is would be 840MB/sec. That’s all in a perfect world of course, but you get the idea.
Of course, the RAID configuration matters and needs to suit the data type you are building for, but in general, with spinning media, you can say that more spindles equals more performance. Yes, there are power usage considerations, and cost considerations (though not a great deal), and those all need to be taken into account, but I’m talking about performance here, so stay focused on that side of the equation.
On the SSD side of the equation, there is in fact a big difference between the performance numbers of a small drive vs those of a large drive. Yes there are cost differences as well, but let’s look at the numbers …
Looking at Intel’s DC3500 SSD (a very, very good product imho), there is not a great deal of difference in the IOPs speed from the 80GB to the 800GB drives (70K to 75K respectively), but in the streaming speed there are some pretty dramatic differences. The 80GB drive claims a sequential read speed of 340MB/sec, while the 800GB drives claims a sequential read speed of 500MB/sec. The write speed difference is even more dramatic, with the 80GB drive writing at a claimed 100MB/sec, and the 800GB drive writing at a claimed 450MB/sec.
So why does this matter?
As in any RAID array, more spindles (or in this case drives) matters. If the controller card can split the reads or writes across multiple drives then it reads or writes less data to each drive, finishing tasks quicker. If you add to this the fact that the larger drives are dramatically quicker, then the effect is multiplied.
The conundrum here is that a small number of drives will still suffer from the RAID limitation that the reads and writes to the drives will be larger than it would be if they are spread out across a lot more drives, so a balance is required.
8 x 80GB drives in RAID 0 would give a potential read speed of 2720MB/sec at 640GB capacity.
5 x 240GB drives would give pretty much the same speed, at 1.2TB capacity
So you can’t do a straight upgrade of drive size to come up with the same capacity numbers and still maintain performance – the mathematics of RAID still means more drives equals more speed, but you start to see the point.
The real problem here lies in performance testing, and this is what brought this to my attention. A lot of people don’t read the fine print regarding the speed of the drives – they just look at the marketing blurb and see that the Intel drives are capable of “up to” 500MB/sec read speed (that’s pretty much the way I read the marketing material as well).
So when testing 8 x DC3500 SSDs, I should be able to get 4000MB/sec read speed in RAID0, correct? The answer to that is “yes … if you are using 240GB or larger drives”. However if you only have 80GB or 120GB drives, then your maximum speeds will be 2720MB/sec and 3560MB/sec respectively (in a perfect world).
This will have you yelling at your RAID vendor that their performance is not good enough … it’s holding back the drives! Whereas in reality it’s the drives holding back the controller. Adaptec’s 7 and 8 Series controllers can stream data to the full extent of the PCIe bus (approx 6600MB/sec) … if you put enough of the right speed SSDs on the card. However even if you put 16 x 80GB Intel DC3500 drives on your controller you’ll be shy of that performance benchmark simply because the drives are not as fast as you might have thought.
All of this matters because? Are people really using pure SSD systems now? Well yes, they are, in ever-increasing numbers. The SSD has come of age in the mind of the enterprise, small business and consumer – they are now quite rightly regarded as reliable, fast and reasonably-affordable devices that make such a dramatic difference to a computer system that they are at least worth considering.
Just make sure that you are configuring them correctly, and that you are in fact sure of what speed you are actually buying. This has not been a bash against Intel – all SSD from all vendors suffer the same configuration issues – so in fact “size does matter” – sorry Steffi :-).