Forum Discussion
johny2010_1
15 years agoContributor
Hi Alex,
Thank you for the valuable information you gave.Since you are an experienced user and a volunteer to help, I admire this.
Can you give some more advice for a newbie like me ? Have you done load tests before ?
And if so, at what points you have found problems ?, and how did you solve them ?
By the way, you wrote that crashing is a serious problem but not a performance decrease, and I think you are right,
but actually it is not me but help calls it crash.
Actually , I do not mean that a server can not operate in any way by using the word "crash"
For this situation, what I mean by crash isthat, if a server crashes , either it gives respond more slowly or no respond at al, in this becuuse it is under heavy load.
you wrote
"For example, the slow-down from 1 minute to 3 minutes when the load increases from one user to two users"
you mean 1-3 minute, total test time , which also includes times for helper operations ?
Let say that it is more user in my case, and not a repot case but a http load scenerio
The test complete time increasees, but process performance not changes that much.
At this point, how performace is important ?
Another question is;
If I found decreases for some reason, this means problem somewhere configuration, application code
or this is the real and final capacity for that server for that kind application with that number of users ?
Do I have to investiagte all problematic request to find the real cause ? And how ? It juast gives warning message lets say, how can I know which part is problematic in the whole system ?
How can I learn real capacity for a server machine ? Which files I have to investigate ? Which configurations I have to know ?
Thank you for reading this long post.
Thank you for the valuable information you gave.Since you are an experienced user and a volunteer to help, I admire this.
Can you give some more advice for a newbie like me ? Have you done load tests before ?
And if so, at what points you have found problems ?, and how did you solve them ?
By the way, you wrote that crashing is a serious problem but not a performance decrease, and I think you are right,
but actually it is not me but help calls it crash.
Actually , I do not mean that a server can not operate in any way by using the word "crash"
For this situation, what I mean by crash isthat, if a server crashes , either it gives respond more slowly or no respond at al, in this becuuse it is under heavy load.
you wrote
"For example, the slow-down from 1 minute to 3 minutes when the load increases from one user to two users"
you mean 1-3 minute, total test time , which also includes times for helper operations ?
Let say that it is more user in my case, and not a repot case but a http load scenerio
The test complete time increasees, but process performance not changes that much.
At this point, how performace is important ?
Another question is;
If I found decreases for some reason, this means problem somewhere configuration, application code
or this is the real and final capacity for that server for that kind application with that number of users ?
Do I have to investiagte all problematic request to find the real cause ? And how ? It juast gives warning message lets say, how can I know which part is problematic in the whole system ?
How can I learn real capacity for a server machine ? Which files I have to investigate ? Which configurations I have to know ?
Thank you for reading this long post.