![]() |
![]() |
![]() |
||||
Welcome to the GoFuckYourself.com - Adult Webmaster Forum forums. You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join our community today! If you have any problems with the registration process or your account login, please contact us. |
![]() ![]() |
|
Discuss what's fucking going on, and which programs are best and worst. One-time "program" announcements from "established" webmasters are allowed. |
|
Thread Tools |
![]() |
#1 |
Confirmed User
Join Date: Jan 2001
Posts: 3,539
|
load averages
load average: 1.74, 4.15, 3.18
That's mine. Is this about average? What is yours? Can anyone give me some background on load averages? |
![]() |
![]() ![]() ![]() ![]() ![]() |
![]() |
#2 |
So Fucking Banned
Join Date: Mar 2002
Location: Far out in the uncharted backwaters of the unfashionable end of the Western Spiral arm of the Galaxy
Posts: 893
|
In short and simple words..
Load average is the number of processes waiting for processing each clock cycle (CPU). The first number is 1 minute average, the second is 5 minute average and the last is 15 minutes average. In general it is basically advised to have less that 1 per-cpu - so if you have a 2 cpu server, its advisable to have a load average of less than 2 (especially in the long run). More than 1 per cpu means that there are processes that are waiting in queue to be processed and that is "inefficient". BUT I would add this - what matters is the response time of your servers - NOT the load average. Load average may be an indication of load problem but I know servers who run very well with load average of 6 per cpu and even more. Measure your key requests response time - and if you are satistied, its good enough. Your numbers seems ok though notice your 1 minute is lower than the 5 - which means "now" its better than the average (maybe your server was busy doing something). If you have 2 CPU its good enough. If 1 cpu - can be better.. Trying to solve load problems and reduce load average is always a good thing but you have to know what it is that raises the load average. ![]() |
![]() |
![]() ![]() ![]() ![]() ![]() |
![]() |
#3 |
Confirmed User
Join Date: May 2002
Location: New York City
Posts: 403
|
It also depends a lot on software design.
Some software pushes load averages much harder than others, while accomplishing the same task in the same time. I have seen some webservers at a 100+ load average running apache, fairly efficiently, move to thttpd, with a load average of 0.5, and still accomplish the exact same goal. Load average is not a good way of checking server health, in and of itself. |
![]() |
![]() ![]() ![]() ![]() ![]() |
![]() |
#4 | |
Confirmed User
Join Date: Sep 2001
Location: Kentucky USA
Posts: 689
|
Quote:
if you have perl ad rotator scripts get rid of them and try a php ads rotator if you want to cut down on load and zombies Brian |
|
![]() |
![]() ![]() ![]() ![]() ![]() |
![]() |
#5 |
Confirmed User
Join Date: Apr 2002
Location: San Francisco
Posts: 851
|
About a teaspoon and a half.....
|
![]() |
![]() ![]() ![]() ![]() ![]() |