Messages in this thread |  | | Date | Mon, 04 Feb 2002 22:01:50 +0200 | From | Jussi Laako <> | Subject | Re: [PATCH] improving O(1)-J9 in heavily threaded situations |
| |
Ed Tomlinson wrote: > > One point that seems to get missed is that a group of java threads, > posix threads or sometimes forked processes combine to make an > application. Linux, at the scheduler level at least, does not have the > ability to determine that all the tasks are really one application. > Under light loads this makes no difference. When the load gets heavy > having this ability helps here.
My application is very good example of this kind of application. I'm very worried about the way new scheduler is beginning to behave. It's combination of single processes with many threads and many processes with single threads. And because it's kind of realtime application, _all_ processes and threads are "interactive". CPU load is typically very high.
- Jussi Laako
-- PGP key fingerprint: 161D 6FED 6A92 39E2 EB5B 39DD A4DE 63EB C216 1E4B Available at PGP keyservers
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |