-- [ Picked text/plain from multipart/alternative ] There's a couple of different top versions/outputs depending on dist/version so be a bit wary. I've seen a couple where the max was 50% or 100% depending on which one.
First thing I tend to do is run something like prime95 or something else that will max a virtual cpu and see what it shows. That will give a base 100% virtual cpu usage line for you to play around and check what makes sense thereafter. Be also wary about dividing cpu usage / 2, as whichever way you look at it, it doesn't really work like that. If its a later version of top and you hit "1" it will show separate cpu states which you may find more useful, and is per virtual cpu. I'm thinking that chances are it will still show 14% (as said there's a couple of different tops, so I don't want to say which for sure), and in reality you are using 14% of a real cpu, but you have some "spare" processing time depending on what else is running. Typically you won't get much past an extra 10/15%, probably be even less with gameservers. Main thing is not to think of each virtual processor as a full cpu, even though either one can run at 100% in a sense, but not both at the same time, other things are happening. It's worth doing a bit of reading on hyperthreading on the net which can explain the details a lot better. Best stats are fps and playing and mrtg whines per hour ;). On 9/21/06, Hoder Balder Marcus Jensen <[EMAIL PROTECTED]> wrote: > > Thanks for the answers, looks like it would be best to use a SMP kernel > with HT in bios, since I'm running multiple servers on the machine, now > time for some testing. > > On Thursday 21 September 2006 11:44, *Rüdiger Meier *wrote: > > > So if you run a single threaded process it can show you only 50% CPU > > usuage (=100% of one CPU!) > > > > If you benchmark a single threaded process you will see that it got > > almost the same speed with or without HT activated (amlost - because > > smp got a litle bit overhead). > > > > If you benchmark 2 single threaded processes at the same time you > > should see a bit advantage with HT activated (not the double because HT > > CPU is not a full dual core!) > > > So let me make sure i get this 100% :) If I'm running eg a single > threaded process, like a cs:s server and the top command shows it's > using 14% at the process. eg.: > > 14255 hlserver 15 0 165m 64m 12m S 14.7 3.2 8:57.10 srcds_i686 > - where 14,7% is the cpu usage, it's actaully only using 7% of the CPU, > because thats what I'm seeing > > Any suggestions to a benchmark "program" btw., other then running bots > in cs:s :) > > > _______________________________________________ > To unsubscribe, edit your list preferences, or view the list archives, > please visit: > http://list.valvesoftware.com/mailman/listinfo/hlds_linux > -- _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
