Storwize embiggened: Nearly 4PB of raw capacity, croons IBM's Barry Whyte IBM has substantially refreshed its SVC (SAN Volume Controller) and Storwize storage array products, providing doubled processing power, boosting compression power as well as increasing Storwize capacity to around 4PB, it claims. There are two detailed blogs about the new hardware and software: one from Master Inventor Barry …
Wow, imagine that!
Seriously, that's been a missing part of the formula for a while. I believe the technology existed in the DS8k series for a while but was sorely needed in the midrange. As soon as EMC, HP, or Dell heard storage tiering was a consideration in an RFP they would shred the Storwize on the lack of 3 tiers.
The thing of it was all 3 were doing it and claiming success, meanwhile only IBM was doing 2 tiers and claiming similar capacity efficiency through real time compression on fast disks. That made IBM the odd-man-out in competitive bids where tiering was important and I know it cost them business locally.
Having three tiers is mostly insignificant. Yes, there are certain circumstances where a single volume may have very important active data and the occasional bit you never want to access, but for most people if that data was at one point so critical that it demanded flash storage, it will still be important enough once demoted to warrant something better than a 4 TB 7.2K drive. For VDIs it can be a very big deal, but even with those the host admin's already going to separate out the data. Still, it's been a long time coming and way overdue.
For those of us not interested in vendor top trumps and salesman bullshit, the auto-rebalancing part is way more significant. If IBM can extend this to automatically compressing data which will benefit from it as part of this auto-rebalancing then it really will have something to take to the competition.
Hats off for quoting a non-cache performance figure. Nice change from "We can get a zillion IOPs in a pointless test in a lab environment on ssd storage". A cache miss is a cache miss and the most important factor for comparing storage products, at least from a performance point of view. On which point I'm not sure why they've gone with an SSD shelf instead of one of their flashsystems.
The point is you can do either (FlashSystem is supported behind Storwize family controllers) or both (or neither and choose to virtualise a Violin system or other 3rd party flash or SSD based system).
FlashSystem is great where low latency is required, but if you only need a few TB of "flash" you can add the expansion shelf and drop a small number of SSDs in.
Pick the optimum solution, rather than being forced to do it the only way the vendor can do it - another example of how being a storage hypervisor gives much more flexibility....
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
