<snip> And? Noticed any significant performance advantage? </snip>
Massive increase in performance on mysql VMs with database sizes that exceed memory size (file caching). Boot times on VMs (windows and linux) under 10 seconds. There is no noticeable change in performance for normal operations on normal VMs because most of the files they're IO blocked by are already cached in memory. I actually went with consumer-grade SSDs (4x OCZ 120gb models) in a raid 10. I know most people say 'those aren't good enough for me'. They are! And as long as you plan for some of them to fail over time, you're still ahead on cost and performance vs enterprise-grade SSDs (read: intel). Synthetic testing with hdparm (sdb is the SSD array, sda is the spinning disk array) is below. This comparison is against 7200rpm disks; I don't have hdparm installed on a box running 15k rpm disks: hdparm -tT --direct /dev/sdb /dev/sdb: Timing O_DIRECT cached reads: 1128 MB in 2.00 seconds = 563.48 MB/sec Timing O_DIRECT disk reads: 1276 MB in 3.00 seconds = 425.01 MB/sec hdparm -tT --direct /dev/sda /dev/sda: Timing O_DIRECT cached reads: 138 MB in 2.03 seconds = 68.03 MB/sec Timing O_DIRECT disk reads: 364 MB in 3.00 seconds = 121.32 MB/sec _______________________________________________ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
