chris at imvu.com
Thu Apr 19 18:56:05 UTC 2007
We have seen latency on the memcached server end in one scenario: large
multi-gets from other clients. Due to a bug part of our application was
occasionally generating a 10,000+ key multi-get. This caused other clients
to see pauses while the memcached server handled that request.
Other than that we see no significant latency.
On 4/19/07, Steven Grimm <sgrimm at facebook.com> wrote:
> Lexington Luthor wrote:
> > memcached is certainly not causing your performance problem. Both java
> > and ruby have garbage collectors which can cause huge latency problems
> > particularly if every 30ms matters to you. I don't know about ruby but
> > java's GC can not be turned off in your code, so you will likely have
> > GC-related latency problems. Some VM language implementations e.g.
> > CPython will let you turn off the GC though.
> On a multi-CPU machine, I'd recommend switching to Java's concurrent
> garbage collector, which in most cases basically eliminates long GC
> delays. It does cost more total CPU time than the non-concurrent
> collectors but it makes things much smoother.
> What does CPython do if you turn off GC? Does the process just keep
> getting bigger and bigger until the machine runs out of memory? Python
> used reference-counting internally last I looked; if you turn off
> garbage collection, how does it ever free up objects with circular
> references? Or do you just have to be careful not to create circular
> references in that case?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the memcached