first time user with out of memory question
time at digg.com
Mon Jun 12 22:27:00 UTC 2006
> There is a reason we spent time reworking memcached's network I/O to
> reduce its CPU consumption: we were getting uncomfortably close to
> running out of CPU capacity on our memcached servers (and these are
> not slow old machines).
Oh, now this is extremely interesting.
When you were getting "uncomfortably close" to the CPU max, what would
you guess was the rate of CPU increase vs. the rate of memcached usage
Our memcached nodes utilise 1/50% CPU at 400 hits/sec. At volumes around
400 hits/sec, the CPU:workload (C:W) ratio is <1. Assuming a constant
C:W ratio, this means we should utilise <<20% CPU at 400k hits/sec.
Given you were hitting CPU discomfort levels at 19k hits/sec, C:W must
become >>1 at those volumes. (I doubt our memcached node CPUs are 20x
faster than yours... waaa-ait a minute, maybe Facebook is buying up all
those $100 laptops???)
More information about the memcached