Maths
"Multiply that by 16, and you get 2,880,000 IOPS which is 500,000 less than the 2.3 million predicted by Oakgate"
It's 500,000 *more* last time I checked...
Solid state drive startup Pliant has announced a benchmark saying its SSDs rock - as it said back in September. Why is it bothering to tell us? The benchmark is from an independent tester, Oakgate Technology, but it was presumably paid for by Pliant. It claims 16 Pliant 300GB LS 300S SSDS delivered 1.1 million IOPS "with a …
Pliant has been claiming 180,000 IOPS per SSD in -- MIXED -- workload scenarios.
Some comments...
1) 68,000 IOPS (1.1m divided by 16 ssds) is less than 40% of claimed performance. Still -- seems like a good number, except....
2) ...according to the video, this number was "achieved" by setting queue-depth at 64 outstanding IOs per drive!!
This is a ridiculous queue-depth number that would never be seen in real world applications...never never ever.
Real-world applications are RARELY able to generate more than even 6 or so outstanding requests at any given target, and the average is only 4. The idea that there are 1024 I/Os queued up in this system (64*16) is utterly ludicrous. Likewise the Pliant/Oakgate benchmark result is pure fantasy.
Back in 2008, Pliant slammed a competitor based on their published SPC-1 benchmark performance -- claiming that their SSD would cost 1/4 or less in terms of $/IOP.
http://blog.enterprisestoragesense.com/2008/04/14/never-send-hdd-to-do-the-job-fit-for-efd/
So...I wonder why Pliant didn't go the SPC-1 route? All they need to do is show about $0.25 per IOP on SPC-1 to match their previous performance claims and settle the matter.
Should be easy right?