Hello,
I would love to:
1) Hear what has been your top bandwidth use measured in MB/Hour and
2) How many users were logged on that peaking hour in the Moodle site you administer?
I look forward to see your replies 
Thanks,
Jens
Jens-This is an interesting question, but I'm not quite sure how I could answer it. My stats software shows averages and totals per hour, but not for a single day.
N Hansen - thanks for your attention. Any one else who has such stats?
BR
Jens
I don't think many people keep hourly stats.
But why do you want them? There are so many variables such as network bandwidth, content type, usage patterns, number of users etc .... I can't see how such figures would be useful ...
But why do you want them? There are so many variables such as network bandwidth, content type, usage patterns, number of users etc .... I can't see how such figures would be useful ...
It is useful for me as one of many variables on one side of the server performance issue. For the admins that don't have hourly stats then they canlook for the peaking hour of all their stats and give a short info about the number of users at that time and their respective bandwidth usage on that peaking hour.
Thanks,
BR
Jens
Hmm, I'm not sure how useful this info. would be. For instance I have a few courses where there is alot of audio and video content, and they use more bandwidth than all the other courses combined.
If one teacher had 100 pages of readings in pdf and another 100 pages in html, the bandwidth difference between the two classes would be dramatically different even with the same number of students reading the same number of words at the same time.
So the numbers you'd get from your question would be wildly variable based on factors that have nothing at all to do with Moodle. Moodle itself doesn't use much bandwith, once a student has cached the css and the javascript files, not any more than delivering the same content statically, so you really need to look at the sort of content a specific site is delivering to get an idea of the bandwidth needs.
And conversly, the numbers from one site will have very little in common from the numbers of a second site unless the two sites are delivering the same sort of content mix (html vs. pdf vs. image vs. audio vs. video).
If one teacher had 100 pages of readings in pdf and another 100 pages in html, the bandwidth difference between the two classes would be dramatically different even with the same number of students reading the same number of words at the same time.
So the numbers you'd get from your question would be wildly variable based on factors that have nothing at all to do with Moodle. Moodle itself doesn't use much bandwith, once a student has cached the css and the javascript files, not any more than delivering the same content statically, so you really need to look at the sort of content a specific site is delivering to get an idea of the bandwidth needs.
And conversly, the numbers from one site will have very little in common from the numbers of a second site unless the two sites are delivering the same sort of content mix (html vs. pdf vs. image vs. audio vs. video).
I'm with MD and Michael here. It is a pretty meaningless set of stats. And you can sure count me as a stats/profiling freak.
There is only one scenario where it does matter: for the very bandwidth constrained, or for those paying premium by traffic. In both cases, all you can do is avoid heavy media and monitor your own traffic (or read the stats that the ISP provides -- if they charge you by traffic, they'll give you good stats).
BTW, if you really need those stats because your bandwidth is expensive or limited, then things like webalizer aren't good enough. You need the stats from the router/firewall (or at least from your tcp/ip stack). Webalizer can see only served Moodle traffic -- won't account for emails, file uploads, or anything other than the data that Apache pumped out.
Hello!
I am also with Martin (the both of you ) and Martín thanks for your extra information on how to the retrieve that information that I really need also if it only approximated values. I hope it can help a few stats out in the open.
BR
Jens
We organise classes for teachers simultane on different locations. During these meetings they are teacher in a course. And then Moodle is very, very slow...
(Book? Quizz? Forum?? Built-in HTML-editor???...)
When we start a conversation with the sysadmins, it is very difficult to find clues for possible answers:
- Apache? (number of concurrent users?...) Mysql? cache? (ok) network? (100 Mbps) Moodle? Special parts of Moodle?? (like the not so clever way of the forum with all it's attached resources, the book that invites you to look up quick several pages (with attchments, what happens when you goto another page duing an unfinished query... etc...)
SO:
- What are the golden rules for tuning a LAMP
- How can we observe which part of Moodle is causing the jam.
- (Or prove that it is NOT the cause
)
> How can we observe which part of Moodle is causing the jam.
This is what I recommend you do. Your sysadmins should know how to deal with diagnosing performance and bottlenecks on the server. I've posted quite bit on this forum about tools to do that. The forum search tool has gotten so good, try it
This is what I recommend you do. Your sysadmins should know how to deal with diagnosing performance and bottlenecks on the server. I've posted quite bit on this forum about tools to do that. The forum search tool has gotten so good, try it