X-list: oracle-l Return-Path: Subject: Re: Confio DbFlash for Oracle From: DEEDSD@Nationwide.com Message-id: OF5CDC7E5C.DD167569-ON852570ED.005F7EAC@lnotes-gw.ent.nwie.net Date: 2006-01-05 18:28:24 I've been using it for about two years now. We licensed 272 CPUs worth initially, and it paid for itself in the first six months. We have an environment with 7 DBAs and we're closing in on 500 databases. Proactive performance tuning was totally out of the question. We were in complete fire-fighting mode and I was the only DBA that had an interest and aptitude for performance. While I was capable of doing everything by hand, there was only so much of me to go around, and I was constantly buried. Consequently, I could only focus on a very limited number of performance problems at any given time, and they took a fair amount of time to resolve. Given the state of tuning (not much up to the point I got involved) we desperately needed a tool that could monitor things 24x7 and dramatically cut down on the time spent troubleshooting performance problems. I did an evaluation with several such tools, and price & features clearly showed DBFlash as the winning product for us. In the first month after we had purchased the product, we had critical performance issue pop up. Everyone had their eyes on the process, and had 20 resources tied up in troubleshooting meetings for a week or two before I got the call. I installed DBFlash on the database and let it run for 20 minutes, and I had my answer, in lovely, graphical form that everyone involved could see. It turned out the problem was a rare Oracle bug the inundated the database with library cache latch and load lock waits. I found the bug on Metalink (I think we were the third customer in the world that hit the bug) , installed the patch and everyone was happy in less than 4 hours start-to-finish after I was called. When you consider that we had 20 people averaging around $40/hr for a couple of weeks in a constant meeting (with laptops & everything, looking at everything possible 8 hours per day), DBFlash would have saved us $60,000 in productivity had I got the call right away instead everyone not being able to narrow it down to a database issue. This is just one of a whole list of dramatic examples of what you can achieve when you have no idea where the problem is, and you have a tool that shows it to you. -- http://www.freelists.org/webpage/oracle-l