Robert L. Cochran wrote: > I'm new to the CPU-level stuff and don't have much background in them. > So some of the terms are new for me. Let me get really basic: why > would you want multiple processors on a motherboard as opposed to a > single fast processor? Well I personally can think of 3 instances that would call for multiple cpus- 1)You are using an app that is multi-threaded, or one that starts multiple proccesses. -For example an nfs or web server running multiple instance of itself -compiling a kernel with "make -j 4 bzImage" 2)You run 2 cpu intensive apps at once. -For example I use to compile kernels on my workstation, while I was doing email. 3)You have multiple users on a system. > I take it that the Xeon line is for multiple CPU motherboards -- you > don't just run one Xeon, am I right? Well you can run 1 xeon, but the main point of new xeon is that P4's can't do smp. > What does it mean, to be 'cpu cache bound'? A proccessor has an internal cache (also L1, or L2 cache). Accessing this cache is much much faster than accessing ram. Some applications will see greater speed improvements with slower proccessors with 2meg cache, vs proccessors twice as fast with 512 caches. The issue with hyperthreading is that it effectively divides your proccessor cache in 1/2. > > Do your comments also mean the Red Hat kernel won't need testing on > the new Hyper Threaded P4s? Well they will need to test it, but I suspect they won't need to make changes. I'd bet that they have had hyperthreading engineering sample P4s for a couple of months. Also keep in mind that intel motherboards are always tested on 2 OSes Red Hat(something), and MS (something). Intel has a fair number of full time linux hackers. > > Thanks > > Bob Cochran > > Samuel Flory wrote: > >> Red Hat has support this since one of the 7.2 kernel updates. This is >> old hat on the current crop of Xeon (aka P4 Xeon). Linux treats them >> as multiple cpus. Don't assume that this will make your system >> faster. If you tend to only one process active at a time then it will >> slow things down. It's also really bad if you are cpu cache bound. >> >> >> >> > > > -- Psyche-list mailing list Psyche-list@redhat.com https://listman.redhat.com/mailman/listinfo/psyche-list