Re: Help in DSM design

From: Aman Singla (aman@cthulhu.engr.sgi.com)
Date: Mon Mar 06 2000 - 18:06:07 EST


Aman Singla wrote:
>
> > That turns out not to be the case. It was exactly this granularity
> > problem which killed the only machine I'm aware of which did DSM in
> > hardware (the Kendall Square boxes --- and how many people here have
> > heard of that one? :)
>
> The KSR boxes aren't the only h/w DSM systems - NUMA boxes today
> (Origin for one) also do DSM. Both add that additional level of
> locality that a performance conscious application needs to exploit;
> agreed that KSR was the only one to have caching at the remote
> memory to local memory level (with a page size granularity).
> A DSM system could look very much like the current NUMA h/w with
> an added remote-local memory caching happening at application specific
> granularity. The main issue there is getting access control at
> the desired granularity for the s/w DSM layer - which in a way
> the current OS dictates to be a page size; and yeah! unfortunately
> there are many smart solutions around that one that I'm aware of :-)
        ^^^
OOps! meant to say there aren't ....

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:21 EST