You want flash disruption? You got it. Pure Storage's FlashArray is cheaper and faster than enterprise disk drive arrays but just as available and reliable. Cue major oops and headaches in Hopkinton, Sunnyvale, Palo Alto and other storage vendor HQ locations. Pure Storage is coming out of stealth with a product and significant …
Nice To See
Very nice to see Cerberus moving from galactic villainy into data storage.
I was going to have a productive weekend, but now I'm going to be spending it replaying Mass Effect. I hope your happy with yourself!!!
Thanks for the RAID details... Write Life still ignored...
Thanks, Chris, for including the RAID details, with how this vendor is considering dealing with write cycle limitations... multiple levels of independent parity.
Does the flash all get removed via the front panel, with Pure, like traditional disk drives?
With the flash being manufactured at the same time and the array being put under constant write load - I would expect we would still see failures happening simultaneously across the array.
I don't know if I would feel comfortable in replacing long-term storage with high-write loads with something like this. Especially OS-boot disks, as far as whether 2.5" drives will survive. Maybe, flash for the file systems, while logging and paging goes to 2.5" disk.
Until the write life barriers in flash is mitigated, I really don't picture 2.5" disks going away any time soon (unless it is for 1.8" disks, where the space is used to accommodate flash for static data such as OS boot and regular application software while the dynamic data is held on rotating media.)
A solution to flash degradation...
Agreed, except is there really any need for paging/swap files now memory is so cheap and plentiful? Important data and log files could be automatically synchronised to "the cloud".
Multiple Drive failure but...
I can see your point that you will hit a point where drives start failing more and if they are all build together and put under the same load you can expect it to happen all at once. The thing is that it is not likely the same parts of the data array inside a drive will go in all the unit and even if it does the drives should detect the problem before a write and the raid sat ontop of it means that the data should be recoverable from other sectors on other disks? So yes your shiney fast drives will all need pulling and the raid rebuilding after a given amount of time, but if they are half the drive of spinning storage does that matter as long as you only have 80% usage and can start swapping disks in time?
Ripe for an Oracle takeover?
Doing the math -- this doesn't add up...
I'm buying Fibre-channel RAID storage from a tier-one enterprise IT maker, running RAID-10 on 10K rpm disks and paying less than $2/TByte. That's without compression. When I want Flash, I just buy the drives and plug them into the same array and software handles moving the hot data around.
Apart from bogus cost/capacity comparisons, how is there anything new here?
I can take my existing array (on the market since 2009), fill if full of Flash SSDs, turn on dedupe and compression and have the same thing at Purestorage is selling.
Seems that the only "innovation" is that I can't use spinning disks at all in this thing, even when they make better sense.
It's combining 2 benefits: Flash AND deDupe - hence the maths error
If you de-dupe of COURSE you're going to get space efficiency compared to "standard" SAN.
I wonder if the claims add up as well when you layer de-dupe with trad hybrid? You still get best £/TB on 7.2k SAS, and best price/perf on SSD/Flash, so as "any fule no" when you hybrid the two you get the best of both worlds.
If they have simplified the system by going Flash only, and then hiding the £/TB with de-dupe, then yes, it aught to knock spots off trad SANs, but I wonder if it stacks up so well when you include deDupe, such as Dell's new FS7500 filer into the mix.
I also wonder how efficient their in-house de-dupe really is against Block data... ?