On 08/12/13 11:10 PM, Jeff Tan wrote:
> Hi Sébastien
>
>> How many rank are you using per Blue Gene node ?
>
> Appears to have been 4 processes per node on the BG/Q. Is there a safe
> level of distribution per node?
>
Hi,
When I did my tests a while ago, I think I was running at 4 ranks per node
Hi Sébastien
> How many rank are you using per Blue Gene node ?
Appears to have been 4 processes per node on the BG/Q. Is there a safe
level of distribution per node?
Regards
Jeff
--
Jeff Tan
High Performance Computing Specialist
IBM Research Collaboratory for Life Sciences, Melbourne
Phone:
On 02/12/13 07:43 PM, Jeff Tan wrote:
>> In the tests I did on this platform, I found that in general
>> Ray needed between 400 MiB and 1 GiB of RAM per MPI rank.
>
> We must be doing something wrong then, because we were crashing out with
> debug messages indicating over 2 GB in some ranks at leas
> In the tests I did on this platform, I found that in general
> Ray needed between 400 MiB and 1 GiB of RAM per MPI rank.
We must be doing something wrong then, because we were crashing out with
debug messages indicating over 2 GB in some ranks at least:
Rank 130: assembler memory usage: 224757
> > Is there any limitation on the Memory used by Ray for the nodes? What we
> > have found is that
> >apart from Rank 0 all the other Ranks do not used more than 20MB of memory.
>
> Each Blue Gene/Q node has 1 A2 CPU with 16 cores and a total of 64 threads (4
> SMT threads per core).
>
> In t
On 07/11/13 08:27 PM, Munish Mehta wrote:
> Hi Seb,
(Please use the mailing list, to subscribe:
https://lists.sourceforge.net/lists/listinfo/denovoassembler-users )
>
> I am using RAY to do a denovo assembly of Cane Toad (Bufo Marinus).
> We are using IBM Blue Gene/Q system. (65,536 PowerPC base
