Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
CanadaDBA
583 Posts |
 Posted - 2005-05-24 : 11:04:32
|
| On a SQL2000/Win2003 server with 1GB RAM(my test server), I imported data from a text file and ran PerfMon on another server to monitor the process. At the time of import no one was using the database. I have interpreted the results based on the document "How to Perform aSQL Server Performance Audit" by Brad M. McGehee in www.sql-server-performance.com. But there are some misunderestanding matching my data with the document. Here are the results of my PerfMon: 1. The "SQL Server Buffer: Buffer Cache Hit Ratio" counter has:Ave 99.5Min 86.7Max 99.89How we can interpret the Ave. high buffer cache hit ratio? Note that my work was only import.2. The "Physical Disk: % Disk time" counter has:Ave 135Min 0Max 23430What does the 135 or 23430 mean? How a disk can be busy more than 100% of the time?3. The “Memory: Pages/sec” counter has:Ave 21Min 0Max 2023Is Ave. of 21 acceptable? What are the acceptable limits for this measure? Is there any official documentation for this? During the import there are two 5 minutes spikes with more than 100 pages/sec. Can I call it bottleneck?4. The “Processor: % Processor Time” counter has: Ave 7.45Min 0Max 25.55If earlier counters show a bottleneck, shouldn’t this counter be higher? Shouldn’t the processor spend time idling waiting on IO?Canada DBA |
|
|
CanadaDBA
583 Posts |
|
|
|
|
|
Interpreting import monitoring results