Open hofmannc opened 8 years ago
Please evaluate the following test setting scenarios in what way they are adequate for our purposes to create a test environment as close as possible to the live server.
24 processors 2,6 GigaHertz memory 132 Gigabyte
info Sarah: at the moment it is not for sure if srv11 will be used for INGe. Maybe VM, but this decision cannot be made in the next weeks.
-> Cut down the characteristics of the live server in an appropriate way and try to close the gap by quantitative methods
info Sarah: srv 01 (it-testserver) and srv 02 (qa-server) are quite close to the live setting, both: 8 processors 2,83 GigaHertz memory 33 Gigabyte
--> Please find out on -general best practises cutting down of hardware ressources for load test purposes. -what to keep in mind for the cloning of VMs for load test purposes. -Expenditure to implement the approaches
--> please find out on -Adequate tools available? -Advantages? -Disadvantages? -Expenditure for implementation?
So far I can tell the server performances of Live and QA are actually similar, and I don't think the difference in memory would be an issue. The number of cores is more complicated, because we have to find out how much the code utilises concurrency.
postponed since we do not know the actual live server environment
Right now we know that Jmeter is likely be a good tool for us to implement load tests. Next step from technical point of view would be to find out about an adequate test environment: