Hi, IOPS on the OSTs? Your OSTs should probably be optimized for capacity and throughput, not IOPS. Nevertheless, how about Intel Manager for Lustre*, inside Intel Enterprise Edition of Lustre*? Others tools include Collectl, LMT, or "llstat" - the last of these is already on your OSS's. -- Brett Lee Sr. Systems Engineer Intel High Performance Data Division From: hpdd-discuss-bounces(a)lists.01.org [mailto:hpdd-discuss-bounces@lists.01.org] On Behalf Of Singhal, Upanshu Sent: Thursday, September 26, 2013 7:24 AM To: hpdd-discuss(a)lists.01.org (hpdd-discuss(a)ml01.01.org) Subject: [HPDD-discuss] Tools to collect performance numbers on OSS / OSTs Hello, Can someone please suggest some tools to collect performance numbers like IOPS on OSTs while running load from Lustre Clients? Thanks, -Upanshu Upanshu Singhal EMC Data Storage Systems, Bangalore, India. Phone: 91-80-67375604

Thanks a lot Cliff. It seems that IML is not readily available for download, we may need to go with Intel on this. Thanks, -Upanshu -----Original Message----- From: White, Cliff [mailto:cliff.white@intel.com] Sent: Thursday, September 26, 2013 7:34 PM To: Lee, Brett; Singhal, Upanshu; hpdd-discuss(a)lists.01.org (hpdd-discuss(a)ml01.01.org) Subject: Re: [HPDD-discuss] Tools to collect performance numbers on OSS / OSTs From: <Lee>, Brett Lee <brett.lee@intel.com<mailto:brett.lee@intel.com>> Date: Thursday, September 26, 2013 6:41 AM To: "Singhal, Upanshu" <upanshu.singhal@emc.com<mailto:upanshu.singhal@emc.com>>, "hpdd-discuss@lists.01.org<mailto:hpdd-discuss@lists.01.org> (hpdd-discuss@ml01.01.org<mailto:hpdd-discuss@ml01.01.org>)" <hpdd-discuss@ml01.01.org<mailto:hpdd-discuss@ml01.01.org>> Subject: Re: [HPDD-discuss] Tools to collect performance numbers on OSS / OSTs Hi, IOPS on the OSTs? Your OSTs should probably be optimized for capacity and throughput, not IOPS. Nevertheless, how about Intel Manager for Lustre*, inside Intel Enterprise Edition of Lustre*? Others tools include Collectl, LMT, or "llstat" - the last of these is already on your OSS's. -- Brett Lee Sr. Systems Engineer Intel High Performance Data Division From: hpdd-discuss-bounces@lists.01.org<mailto:hpdd-discuss-bounces@lists.01.org> [mailto:hpdd-discuss-bounces@lists.01.org] On Behalf Of Singhal, Upanshu Sent: Thursday, September 26, 2013 7:24 AM To: hpdd-discuss@lists.01.org<mailto:hpdd-discuss@lists.01.org> (hpdd-discuss@ml01.01.org<mailto:hpdd-discuss@ml01.01.org>) Subject: [HPDD-discuss] Tools to collect performance numbers on OSS / OSTs Hello, Can someone please suggest some tools to collect performance numbers like IOPS on OSTs while running load from Lustre Clients? First, the standard *nix tools, such as iostat and vmstat should be fine on the OSS nodes, depending on how LUNs map to OSTs. Lctl get_param obdfilter.*.brw_stats will also supply much useful information although not IOPS directly. Llstat will likely not show you much for IOPS, but again will supply some data. Collectctl is very scriptable, command line style. LMT and IML are GUI tools and quite featurefull . Cliffw Thanks, -Upanshu Upanshu Singhal EMC Data Storage Systems, Bangalore, India. Phone: 91-80-67375604