search for: 410k

...ximum of about 500K/sec. The effect of different buffer sizes is as follows: Size Speed in K/sec, 32MB file of SO_[RCV,SND]BUF Written by TESTNET, 500 65K blocks 16384 50K/sec (3.4-STABLE) 32768 97K/sec (3.4-STABLE) 65535 450K/sec (3.4-STABLE) 131072 410K/sec (3.4-STABLE) 262144 1980K/sec (3.3-RELEASE)

So I got dashboard up and running on our production system on Thursday before I left. Within 48 hours it had completed filled the /var filesystem. The ibdata1 file is currently at 8GB in size. 1. What size should I expect for ~500 nodes reporting every 30 minutes? 2. Are there some database cleanup scripts which I have managed to overlook that need to be run? -- Jo Rhett Net Consonance :

...Q.pdf lrwxrwxrwx 1 root root 41 2005-05-03 12:20 R-intro.html -> ../../../../lib/R/doc/manual/R-intro.html -rw-r--r-- 1 root root 647K 2005-02-21 02:34 R-intro.pdf lrwxrwxrwx 1 root root 40 2005-05-03 12:20 R-lang.html -> ../../../../lib/R/doc/manual/R-lang.html -rw-r--r-- 1 root root 410K 2005-02-21 02:34 R-lang.pdf This links point to them selves. Should I report this to http://bugs.debian.org/r-doc-html? I appologize for wrapping. -- Lep pozdrav / With regards, Gregor Gorjanc ---------------------------------------------------------------------- University of Ljubljana Bi...

...25% write / 75% read | 75% write / 25% read -----------------|------|----------------------|--------------------- 1x rd_mcp LUN | 8 | ~155K IOPs | ~145K IOPs 16x rd_mcp LUNs | 16 | ~315K IOPs | ~305K IOPs 32x rd_mcp LUNs | 16 | ~425K IOPs | ~410K IOPs The full fio randrw results for the six test cases are attached below. Also, using a workload of fio numjobs > 16 currently makes performance start to fall off pretty sharply regardless of the number of vCPUs.. So running a similar workload with loopback SCSI ports on bare-metal produces...

...25% write / 75% read | 75% write / 25% read -----------------|------|----------------------|--------------------- 1x rd_mcp LUN | 8 | ~155K IOPs | ~145K IOPs 16x rd_mcp LUNs | 16 | ~315K IOPs | ~305K IOPs 32x rd_mcp LUNs | 16 | ~425K IOPs | ~410K IOPs The full fio randrw results for the six test cases are attached below. Also, using a workload of fio numjobs > 16 currently makes performance start to fall off pretty sharply regardless of the number of vCPUs.. So running a similar workload with loopback SCSI ports on bare-metal produces...