Message boards :
Number crunching :
Server problems
Message board moderation
Previous · 1 · 2
Author | Message |
---|---|
Send message Joined: 1 Sep 04 Posts: 55 Credit: 20,907 RAC: 0 |
|
Send message Joined: 1 Sep 04 Posts: 1 Credit: 1,515 RAC: 0 |
Out of curiosity, how many connections can it handle before it starts to break down? Zeeno > We had to disable even message boards for a while, the system was really stuck > up. There is certainly a limit in how many database connections our server can > handle at a time... But there was also bug in the main page which > unnecessarily made database queries each time it was loaded. We fixed that, > and also tuned the database server so now the forums are open again. Let's > hope the system works better now. > > Results and Pending credits pages have been disabled until we see that we have > enough capacity to handle them. > > > Markku Degerholm > LHC@home Admin > |
Send message Joined: 1 Sep 04 Posts: 15 Credit: 233,588 RAC: 0 |
> Point is that those short WU's are needed just as much as the long ones. But > we try to generate a mix of short and long work units such that the average is > good, and we get the short ones crunched as well. But I think it will take a > few more days before our physicists are able to start submitting those longer > jobs. Maybe it is possible to combine a few short WUs (i.e. 10?) into one? And calculate this monstrous WU 10 times, for every subWU? ;-) |
Send message Joined: 3 Sep 04 Posts: 212 Credit: 4,545 RAC: 0 |
> Out of curiosity, how many connections can it handle before it starts to break > down? The problem is that the more there a connections, the longer they take to complete, and the more connections are needed. Now that we system is working pretty nicely, we have 50 simultaneous DB connections and about 100 DB queries executing per second. When the system was really slow, there was 400 connections which were able to do about 20 queries per second. After that we saw it wise to limit number of connections to 100. We think now that bottleneck is amount of RAM and the disk system. After the database cannot be cached into available RAM, disk system performance starts to limit the performance. Most important aspect of disk performance is seek time, actual disk bandwidth is no problem. Now that we have a new server (it should be serving on Monday) we should be able to serve current user base without problems. Markku Degerholm LHC@home Admin |
Send message Joined: 2 Sep 04 Posts: 321 Credit: 10,607 RAC: 0 |
good news markku have a nice sunday ;-) http://www.fs.fed.us/gpnf/volcanocams/msh/images/mshvolcanocam.jpg |
Send message Joined: 1 Sep 04 Posts: 55 Credit: 20,907 RAC: 0 |
|
©2024 CERN