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 |
|
AskSQLTeam
Ask SQLTeam Question
0 Posts |
Posted - 2002-03-26 : 09:11:52
|
| Simon writes "Hi AllIm using performence monitor on Win2000 with sql 7 on a clustered enviornment. We are running some tests with 40 users accessing the system and I want to monitor the effects on the system.These are the measures I have selected to use for Performence monitor ;Memory Pages/SecMemory Available BytesPhysicalDisk % Disc TimePhysicalDisk Avg. Disk Queue Length Processor % Total Processor TimeSQL Server Access Methods Page Splits/secSQL Server Access Methods Full Scans/SecSQL Server Buffer Manager Buffer Cache Hit RatioSQL Server Buffer Manager Cache Size (pages)SQL Server General Statistics User ConnectionsSQL Server General Statistics Log In/SecSQL Server General Statistics Log Out/SecSQL Server Locks Number of Deadlocks/SecSQL Server Locks AVG Wait Time (ms)SQL Server SQL Statistics Batch Requests/SecSQL Server SQL Statistics SQL Comilations/SecI do have a couple of questions 1.Any other obvious measures I've missed from my selection ?2. Are there useful correlations I can draw from the above e.g number of user connections and number of deadlocks ?3. For the following measures which thresholds should I be using e.g > 10 for five minutes is bad /good/indifferent ?!SQL Server Locks Number of Deadlocks/SecSQL Server Locks AVG Wait Time (ms)Many thanks for any help ..realise this is a bit of a longwinded one !!!" |
|
|
|
|
|