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 |
|
ferrethouse
Constraint Violating Yak Guru
352 Posts |
 Posted - 2012-08-16 : 12:14:30
|
| I know the answer to that question is no but I want to gauge the extent that my metrics are of value. I used a tool called SQLStress to determine how many transactions/second various AWS instances could support. I found that the cc2.8xlarge instance could support 12190 transactions per second. I then turned on perfmon on my local production server and checked the SQL Server transactions/second metric. It averages about 1000 transactions/second. But I suspect that the transactions contained in the SQLStress software are less complex than many of the transactions performed by our application so I don't know how much comfort I can derive from the analysis. Your thoughts? |
|
|
jackv
Master Smack Fu Yak Hacker
2179 Posts |
Posted - 2012-08-16 : 15:10:35
|
| Compaing the two different traces is probably not of great value. Do you have any other benchmark figures to compare? You can also use TPC and generate some test dats.Jack Vamvas--------------------http://www.sqlserver-dba.com |
 |
|
|
|
|
|
Are all transactions created equally?