Quoting Chris Wright (chrisw@private): > * Chris Wright (chrisw@private) wrote: > > I'll have some numbers tomorrow. If you'd like to run SELinux that'd > > be quite useful. > > These are just lmbench and kernel build numbers (certainly not the best > for real benchmark numbers, but easy to get a quick view run). This is > just baseline (i.e. default, nothing loaded). Here are some numbers on a 4way x86 - PIII 700Mhz with 1G memory (hmm, highmem not enabled). I should hopefully have a 2way ppc available later today for a pair of runs. dbench and tbench were run 50 times each, kernbench and reaim 10 times each. Results are mean +/- 95% confidence half-interval. Kernel had selinux and capabilities compiled in. A little surprising: kernbench is improved, but dbench and tbench are worse - though within the 95% CI. dbench (throughput, larger is better): original: 357.957780 +/- 3.509188 patched: 351.266820 +/- 4.736168 tbench (throughput, larger is better): original: 38.710270 +/- 0.028970 patched: 38.210506 +/- 0.032954 kernbench (time, smaller is better): original: 91.837000 +/- 0.324471 patched: 91.466000 +/- 0.308797 reaim (#children vs throughput, larger is better): original: 1 48702.197000 1875.223996 3 131411.870000 4497.107969 5 130219.174000 6365.289551 7 162377.027000 3131.071134 9 155432.904000 4964.935291 11 169784.384000 4490.812272 13 164540.169000 3902.652904 15 172983.569000 3149.934591 patched: 1 47525.273000 1509.578035 3 132151.651000 2282.043786 5 131244.291000 5874.212092 7 165629.693000 4646.641230 9 156163.110000 3422.903849 11 170608.526000 4132.988693 13 164863.102000 3664.214481 15 172947.803000 2548.662380 thanks, -serge
This archive was generated by hypermail 2.1.3 : Fri Aug 26 2005 - 06:19:20 PDT