On 03/10/2010 11:06 AM, Tanveer Singh wrote:
So we were thinking that we run a simple monitor software which dumps some
graphs etc., over a period of one week, where we can compare the real time
traffic vs CPU load graphs etc., and which process is memory intensive,
which is CPU intensive blah blah.

I feel you are going down the wrong route.

Reducing load and resource utilisation on a machine is always a waste of time. What you want to do is increase the amount of content you can deliver out of the same resources. Think about it, its not the same thing. And the path you take down the route will almost certainly be different for both of those approaches. Dont forget - its a machine that has a finite resource threshold.

Start by setting up a small machine local to you, that isnt in production and test it for performance ( using something like apachebench, ab, as a start - but use something good like tsung for real performance testing ). Use this as a reference platform, eg if you can get 25% better performance by moving all your static content ( eg. graphics, css, js ) into a tempfs, that will translate to a about the same level[1] boost in production as well. If you have a quad core, perhaps locking down 2 cpu cores and dedicating them for the mysqldb might be a good idea. Similarly, separating the i/o for http and mysqld at the block device level might be something you want to look into.

These are just some generic options you can look at - before you move into app level optimisations. And there are many many things that could be done there as well. Simple things like - are you running the webroots with atime enabled / ensure that db index's are valid and the querylog isnt reporting runtimes greater than a second for most things....

Hope this helps and gets you thinking along a more formalised path.

- KB

[1]: its never really the same since traffic shape and sources never really reflect a test environment - unless you do a lot of work, like replaying user access logs as one of your performance metric sources.

_______________________________________________
Ilugd mailing list
Ilugd@lists.linux-delhi.org
http://frodo.hserus.net/mailman/listinfo/ilugd

Reply via email to