EMC on XtremIO SSD brickup ballsup: Its lifetime downtime is under 3 minutes No DOAs here: EMC’s XTremIO arrays are expected to have less than three minutes downtime in their rated life, with X-Brick component SSDs failing once every five years or so. We have reported that the DOA rate was too high and the co-founder and general manager of EMC's acquired XtremIO business, Ehud Rokach, has written blog …
Re: you’ll need to replace a failed SSD (Due to a non-endurance related failure. …) once every..
So what's the endurance related failure rate? Early SSDs without good load leveling algorithms had fairly short lives under active write use (the worst, very short lives). Presumably these enterprise level drives have great load leveling algorithms and lots of spare chips - how long do they last under a lot of write activity?
In most cases, maintenance contracts for enterprise storage normally include HDD replacements. Unless there's something specific in EMC's contracts for SSDs with respect to total writes, I would expect it to be the same in which case SSD endurance ought to be covered (especially as replacements can be preemptive),
Years ago, I looked after a few Sun E4500 that had external disk shelves. We had an almost weekly appointment with Sun engineers to replace failed disks for about 6 months until we finally got someone who knew about this issue and carefully inserted cardboard into the area enclosure. As we watched he looked up and said "the enclosure rattles the disks to death - this fixes it".
So much for MTBF stats....
"the enclosure rattles the disks to death - this fixes it".
As someone who has to maintain several hundred Tb of disk storage it's blindingly obvious that levels of expertise in preventing disks being rattled to bits vary wildly between makers (at consumer level, forget it! Bling is more important than all else)
Paradoxically one of the very worst large enclosures I've run across is made by Xyratex - who are the kind of people you'd expect to know better (then again perhaps it explains why HDD reliability is pants these days). It kind of explains why they got out of the disk array market.
when did xyratex get out of the disk array business? they still make enclosures for sure, HP uses em(on 3PAR 7000 at least), last I saw IBM used them on V7000, NetApp stopped using them I believe after they absorbed that division from LSI. I imagine a bunch of other folks use Xyratex still too, not sure about HDS/EMC I don't make a point to track Xyratex :)
looks like Xyratex still makes their own arrays for HPC as well.
Lets see, we have:
"...we have shipped hundreds of XtremIO X-Bricks to customers..."
"...less than 32 Seconds of unavailability in a year,..."
"...field-measured to be 922,240 hours, or 105 years."
which suggests they have sold just over 200 units in 6 months, one of which developed a fault.
Not the most statistically reliable of claims.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
