Date: Wed, 26 Apr 2006 07:09:25 +0530 From: "Joseph Koshy" <joseph.koshy@gmail.com> To: "Bill Moran" <wmoran@collaborativefusion.com> Cc: freebsd-performance@freebsd.org Subject: Re: Dual-core CPU vs. very large cache Message-ID: <84dead720604251839t3014389dka036b7d7bbd422c@mail.gmail.com> In-Reply-To: <20060425090739.8470143f.wmoran@collaborativefusion.com> References: <20060425090739.8470143f.wmoran@collaborativefusion.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> Our current Dells have 2M cache, and I'm trying to determine > whether the 8M cache will make a significant difference or > not. Can someone recommend a testing procedure for > determining whether adding cache is worthwhile or not? > I can simulate a test load at any time, but I don't know how > to tell whether the cache is the bottleneck of the CPU or not. You didn't say which version of FreeBSD you are running. If this is 6.X, you could run pmcstat(8) in a system-wide counting mode and measure cache behaviour directly. Briefly, # (compile, boot a kernel with options HWPMC_HOOKS) # kldload hwpmc # pmcstat -w 1 -s dc-misses test-load See pmc(3) and the CPU vendor documentation referenced there for the events your CPU can measure. If your application is CPU bound you may want to check for memory bus behaviour, cache behaviour, cycles the processor was stalled and for poor branch prediction. If you notice anything anomalous, you can use the hwpmc's sampling modes to find out what code is causing the anomaly. -- FreeBSD Volunteer, http://people.freebsd.org/~jkoshy
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?84dead720604251839t3014389dka036b7d7bbd422c>