Erik de Castro Lopo wrote:
- Is there any chance your machine is NUMA (Non Uniform Memory
Architecture)? This basically means that all CPU can access
all the memory in the machine, but that for each processor
some memory is local and fast while other memory is further
away (ie on another card on the other side of some shared
bus) and hence slower? If so, you may find that the point
where you get your slowdown coincides with filling up local
memory and starting on more remote memory.
Hi all! It's me again =P
I am successfully able to implement memory locking but now that I am
able to allocate over 2GB fairly quickly, I get this error while nearing
3GB (remember I have 9GB of RAM):
Out of Memory: Killed process XXXX
OOM kill occurred on an x86_64 NUMA system.
Now from my understanding, all NUMA memory is accessible by all process
regardless of which CPU it's running from...so accessing other memory
areas will be slower, but definitely accessible. I don't know why I'm
running out of memory. I have set all my limits in
/etc/security/limits.h to unlimited and am running as room. Any suggestions?
Carlo
_______________________________________________
coders mailing list
[email protected]
http://lists.slug.org.au/listinfo/coders
