From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 736 invoked by alias); 17 Dec 2005 06:03:57 -0000 Received: (qmail 728 invoked by uid 22791); 17 Dec 2005 06:03:56 -0000 X-Spam-Status: No, hits=-2.3 required=5.0 tests=AWL,BAYES_00 X-Spam-Check-By: sourceware.org Received: from tomts13.bellnexxia.net (HELO tomts13-srv.bellnexxia.net) (209.226.175.34) by sourceware.org (qpsmtpd/0.31) with ESMTP; Sat, 17 Dec 2005 06:03:55 +0000 Received: from krystal.dyndns.org ([70.49.125.187]) by tomts13-srv.bellnexxia.net (InterMail vM.5.01.06.13 201-253-122-130-113-20050324) with ESMTP id <20051217060352.NYOK20927.tomts13-srv.bellnexxia.net@krystal.dyndns.org> for ; Sat, 17 Dec 2005 01:03:52 -0500 Received: from localhost (localhost [127.0.0.1]) (uid 1000) by krystal.dyndns.org with local; Sat, 17 Dec 2005 00:58:51 -0500 id 002E9163.43A3A91B.000023EA Date: Sat, 17 Dec 2005 14:27:00 -0000 From: Mathieu Desnoyers To: ltt-dev@shafik.org, systemtap@sources.redhat.com Subject: Re: LTTng very heavy system usage scenario Message-ID: <20051217055851.GA30401@Krystal> References: <20051214022515.GA1488@Krystal> <20051216010459.GA11668@Krystal> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <20051216010459.GA11668@Krystal> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.4.31-grsec (i686) X-Uptime: 00:52:02 up 50 days, 7:51, 2 users, load average: 0.19, 0.07, 0.01 User-Agent: Mutt/1.5.11 X-Originating-IP: [0] X-IsSubscribed: yes Mailing-List: contact systemtap-help@sourceware.org; run by ezmlm Precedence: bulk List-Subscribe: List-Post: List-Help: , Sender: systemtap-owner@sourceware.org X-SW-Source: 2005-q4/txt/msg00467.txt.bz2 Well, I finally managed to finish my brand new LTTng+LTTV implementation. This third usage scenario has been made with it. I tried something that always turns out to make tracers dead : a loopback ping flood (ping -f 127.0.0.1). Here are the results :-) Mathieu Usage scenario 3 Very heavy system usage (ping flood on the loopback pseudo interface) Tests made on a 3Ghz pentium 4, with hyperthreading enabled. Trace taken with LTTng 0.5.0 No event has been lost while tracing. (units are in seconds) Trace duration : 12.785209485 Total cpu time available (approximately : it's hyperthreading...) : 25.57s Trace size : 176M Trace data rate : 13.76MB/s idle cpu time : 7.297325362s (sum of the two HT cpus) System load : (25.57-7.297325362)/25.57 = 71.5% lttd cpu time (process 4773 in the trace) usermode : 3.613249328 traps : none irqs : 21 (3com 3c940 10/100/1000Base-T) : 0.094693456 0 (timer) : 0.015632051 19 (USB and IDE) : 0.000242789 system calls : unknown (beginning of trace) : 0.000024347 175 (rt_sigprocmask) : 0.352830699 3 (read) : 0.697699884 142 (newselect) : 2.055746440 4 (write) : 1.522248853 54 (ioctl) : 0.000014157 Total cpu time accountable to process : usermode + traps + system calls 8.241813708 ratio of time for write / cpu time used : 45.10% ratio of time for write / cpu time available (doesn't mean much with HT) : 32.23% Overall information on the trace event count 9588836 Event rate : 749994 events per second detail per event type (38 event types) Statistic for 'event_types' : state_dump_facility_load : 10 exit : 3 wait : 17 free : 3 fork : 3 schedchange : 42076 syscall_entry : 2297497 write : 427383 wakeup : 37268 soft_irq_entry : 309692 softirq : 25105 soft_irq_exit : 309692 read : 166746 syscall_exit : 2297497 expired : 258 irq_exit : 170318 packet_in : 259254 irq_entry : 170318 tasklet_entry : 62 tasklet_exit : 62 packet_out : 408715 trap_exit : 2448 call : 386428 recvmsg : 258803 close : 93 trap_entry : 2448 select : 1406394 buf_wait_start : 19 page_alloc : 478581 buf_wait_end : 19 create : 8 sendmsg : 127608 poll : 3449 ioctl : 443 open : 79 set_itimer : 11 page_free : 3 exec : 4 seek : 19 OpenPGP public key: http://krystal.dyndns.org:8080/key/compudj.gpg Key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68