Hi Misi, I Would love to go through the copy of the presentation.....
BR Ankur -----Original Message----- From: Action Request System discussion list(ARSList) [mailto:[EMAIL PROTECTED] On Behalf Of Misi Mladoniczky Sent: 01 February 2007 03:19 AM To: [email protected] Subject: Re: ARLogAnalyzer output Thad, Well, rrrLic is the most popular one. Probably because it is easier to understand than hardcor logfile analysis... Regardless of the tool you use (Notepad works), I would urge everyone to use the logfiles more. Both for performance and troubleshooting issues. You do not need to know that much to make use of them. I did a presentation at UKRUG about performance tuning with logfiles. Please let me know if you want a copy of the presentation. It goes through the important API-calls for performance, and maps them to the actual workflow involved to trigger them. Best Regards - Misi, RRR AB, http://www.rrr.se > Misi, > > I've used your rrrLic tool and guess I forgot about the others. I'll > have to check them out. > > Thank you, > Thad Esser > Remedy Developer > "Argue for your limitations, and sure enough, they're yours."-- > Richard Bach > > > > "Misi Mladoniczky" <[EMAIL PROTECTED]> > Sent by: "Action Request System discussion list(ARSList)" > <[email protected]> > 01/31/2007 06:47 AM > Please respond to > [email protected] > > > To > [email protected] > cc > > Subject > Re: ARLogAnalyzer output > > > > > > > Hi, > > I usually focus on the number of calls with "no gap", compared to the > total number of calls. > > To this end our rrrLog-tool presents a percentage for a gap less than > 1, 0.1, 0.01, 0.001 and 0.0001 seconds. > > http://www.rrr.se/tmp/rrrLogThread.html > > You have to decide what "no gap" means. My experience tell me to look > at the "less than 0.01 seconds" column. > > To get the exact number of times the server has queued calls, Server > Statistics would be the way to go. > > Best Regards - Misi, RRR AB, http://www.rrr.se > >> I've spent countless hours reading and analyzing files like this and >> troubleshooting performance problems for the last couple years with more >> than 70% increase in performance. In my case the AVG Idle time was >> less than a second and in some cases very close to zero. We ended up >> doubling the number of threads, twice and saw a dramatic improvement. >> While this may be subjective, I would increase your threads, but >> without knowing how many you already have It's hard to say by how >> much. Though I'll stick my neck out and say, increase by about a >> third and evaluate your hardware performance during system load to >> determine whether it > will >> remain that much, AND look at new logs again, then adjust from there. >> By looking at your numbers too, I might suggest increasing Fast >> threads > more >> than List. >> >> This is however a subjective and difficult to say how much better your >> system will be without just doing it and by trial and error. If you >> would like more info, or if I can further answer questions, contact me. >> >> Thanks, >> Rem >> >> >> Date: Tue, 30 Jan 2007 11:16:59 -0800From: [EMAIL PROTECTED]: >> ARLogAnalyzer outputTo: [EMAIL PROTECTED] Hello, I haven't had >> much occasion to use the log analyzer very much, and I want to make >> sure I'm interpreting the results correctly as part of thread tuning. >> In determining if I need to increase my threads, which column is more >> important, the "AVG Idle Time" or the "MIN Idle Time"? I can see it >> either way and was hoping someone would share their experience. Thanks, >> Thad API STATISTICS BY THREAD IDQueue Thread Count Gaps > MIN >> Idle Time MAX Idle Time AVG Idle Time Total Idle Time---------- >> ---------- ------- ------- -------------- -------------- -------------- >> ---------------Admin 001772 83 83 0.0150 >> 46:01.6220 55.6344 1:16:57.6600Fast 000744 4,015 >> 4,014 0.0150 17.8430 1.1034 1:13:49.3890Fast >> 001432 3,977 3,976 0.0160 19.1870 1.1128 >> 1:13:44.7590Fast 001784 3,988 3,987 0.0160 >> 12.8110 1.1131 1:13:57.9920Fast 001956 3,868 >> 3,867 0.0000 13.4830 1.1428 1:13:39.5520Fast >> 002264 3,970 3,969 0.0160 12.8280 1.1217 >> 1:14:12.0360Fast 003660 3,983 3,982 0.0150 >> 19.5150 1.1154 1:14:01.7290Fast 003832 3,950 >> 3,949 0.0160 12.8120 1.1145 1:13:21.3690List >> 001284 1,478 1,477 0.0310 19.9840 2.6610 >> 1:05:30.3070List 001508 1,684 1,683 0.0000 >> 14.8270 2.4314 1:08:12.1150List 001580 1,384 >> 1,383 0.0620 14.4050 2.6266 1:00:32.6020List >> 001616 1,523 1,522 0.0000 16.1870 2.4974 >> 1:03:21.1640List 001732 1,519 1,518 0.0000 >> 13.9830 2.5307 1:04:01.7340List 001752 1,454 >> 1,453 0.0000 13.0150 2.5978 1:02:54.6550List >> 001924 1,499 1,498 0.0000 14.2340 2.5749 >> 1:04:17.3310List 001952 1,565 1,564 0.0000 >> 14.4370 2.4932 1:04:59.4360 >> ***IMPORTANT NOTICE: This communication, including any attachment, >> contains information that may be confidential or privileged, and is >> intended solely for the entity or individual to whom it is addressed. >> If you are not the intended recipient, you should delete this message >> and > are >> hereby notified that any disclosure, copying, or distribution of this >> message is strictly prohibited. Nothing in this email, including any >> attachment, is intended to be a legally binding >> signature.***__20060125_______________________This posting was >> submitted with HTML in it___ >> _________________________________________________________________ >> Get connected - Use your Hotmail address to sign into Windows Live >> Messenger now. >> http://get.live.com/messenger/overview >> > ______________________________________________________________________ > _________ >> UNSUBSCRIBE or access ARSlist Archives at www.arslist.org >> ARSlist:"Where the Answers Are" >> > > ______________________________________________________________________ > _________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org > ARSlist:"Where the Answers Are" > > > > ***IMPORTANT NOTICE: This communication, including any attachment, > contains information that may be confidential or privileged, and is > intended solely for the entity or individual to whom it is addressed. > If you are not the intended recipient, you should delete this message > and are hereby notified that any disclosure, copying, or distribution > of this message is strictly prohibited. Nothing in this email, > including any attachment, is intended to be a legally binding > signature.*** > > ______________________________________________________________________ > _________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org > ARSlist:"Where the Answers Are" > ________________________________________________________________________ _______ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org ARSlist:"Where the Answers Are" _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org ARSlist:"Where the Answers Are"
