I am currently hosting a test website under a 8 vcpu and 2gb ram virtual machine using nginx and php-fpm, the host machine is hosting other virtual machines doing the same thing with 10gbps network interface. During my stress test after 50 users requesting concurrently my test website's response time rises from 800-900ms on normal to around 2 seconds. After this test I tried to increase the virtual machine's ram to 6gb but there were no change in the response time at all. What could be causing this?
Asked
Active
Viewed 769 times
-3
-
Welcome to Server Fault! Your question appears to be a duplicate of [How do you do load testing and capacity planning for web sites?](http://serverfault.com/questions/350454/how-do-you-do-load-testing-and-capacity-planning-for-web-sites) Please review the question and answers there and consider deleting your question. And don't forget to take the [site tour](http://serverfault.com/tour). – Paul Nov 16 '15 at 00:13
-
If you already have enough memory for the workload then nothing will change as you add more memory. – kasperd Nov 25 '15 at 12:04
-
Possible duplicate of [Can you help me with my capacity planning?](http://serverfault.com/questions/384686/can-you-help-me-with-my-capacity-planning) – kasperd Nov 25 '15 at 12:05
1 Answers
4
Oooh the list is kind of endless
- Poor application design
- Poor database queries.
- Poor choice of database engine.
- Sunspots.
- Poor nginx tuning.
- Poor php tuning.
- Insufficient memory
- Swapping.
- etc
So with the information you provided it could literally be anything.
Go away, look at your system using tools like top, vmstat, iostat, sar, free etc. Perhaps even consider a full monitoring solutin Watch what happens to the system resources see which ones are being consumed and take appropriate actions. Rinse and repeat.
It really is science, not lick your finger and hold it in the air stuff. Bring scientific method to bear, don't scream HALP ! at the internet and expect it to guess whats wrong.
