[Beowulf] Storage memory

[Joe Landman]

On 10/31/2013 11:35 AM, John Hearns wrote:
> http://www.theregister.co.uk/2013/10/31/dimm_outlook_for_storage_memory/
> A concept well worth discussing.
> I think I've broached this topic a couple of times on this list - we 
> are seeing ever more applications hungry for large amounts of RAM,
> whether or not that is just programmers dimensioning larger and larger 
> arrays, or applications using finer and finer mesh sizes to
> capture detail of physical processes.
> But does that data have to be in main memory - I think 'storage 
> memory' like this has a great future.
> Also add in the requirement of Big Data in-memory databases too.
> Class, discuss.

:)

The positives are a much larger "randomly addressable" memory pool. The 
negatives are that its still flash (for the moment), and has all the 
issues associated with flash (random small writes == very bad).

For read heavy workloads (in memory databases, etc.) its wonderful. For 
write heavy workloads, you'd want to make sure the devices were hot 
swappable (DIMMs aren't except in specific cases).



-- 
Come by and visit us at SC13 in booth #1919

Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics, Inc.
email: landman at scalableinformatics.com
web  : http://scalableinformatics.com
twtr : @scalablinfo
phone: +1 734 786 8423 x121
cell : +1 734 612 4615