[Beowulf] SSD caching for parallel filesystems

[Vincent Diepeveen]

SSD's are not about bandwidth, they're about latency.

With a raid array of cheapo disks we can also get 3GB/s bandwidth,
more than most 2 socket nodes effectively can handle.

Only theoretically a higher bandwidth will be possible (benchmarks huh).

However getting 20 bytes from a SSD is in the few dozens of  
microseconds, versus several milliseconds for the cheapskate disks.

That factor of 50-100 difference roughly in latency difference is the  
reason SSD's exist.

Any bandwidth test of a SSD is total nonsense.

On Feb 9, 2013, at 10:32 PM, Bill Broadley wrote:

> On 02/09/2013 01:22 PM, Vincent Diepeveen wrote:
>> SATA is very bad protocol for SSD's.
>>
>> SSD's allows perfectly parallel stores and writes, SATA doesn't.
>> So SATA really limits the SSD's true performance.
>
> SSDs and controllers often support NCQ which allows multiple  
> outstanding
> requests.  Not sure if that fits the definition of "perfectly  
> parallel",
> but it does allow for outstanding requests to be answered in the order
> of the devices choosing.
>
> In fact the higher IOP rates would be impossible without it.  Note how
> the latency for a single IOP is much higher than peak 1 second / (peak
> IOPs).
>
>
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin  
> Computing
> To change your subscription (digest mode or unsubscribe) visit  
> http://www.beowulf.org/mailman/listinfo/beowulf