TCP speed versus busy processes
TCP speed versus busy processes
- Subject: TCP speed versus busy processes
- From: "Philip D. Wasson" <email@hidden>
- Date: Mon, 18 Aug 2003 18:45:58 -0400
I have a written a client/server system in which messages are exchanged
over TCP. On Mac OS X it uses the BSD Sockets API, and normal recv(),
send(), select(), that sort of thing. When running the client on my
dual-processor 450 MHz G4 and the server on my 800 MHz Pentium III
(connected via a 100 Mbps switch), I can get 1500 (or thereabouts)
request/response exchanges (unless I measured wrong) over an
established connection. However, if SETI@Home is running on the Mac (on
both processors) at "nice" level 19, the transaction rate drops into
the tens per second. Now, I would think that with my client at nice
level 0 and SETI at nice 19, SETI really should have the CPU taken away
from it as soon as e.g my recv operation is complete, so it shouldn't
slow my client down much. But it does. Is that just due to scheduling
overhead and there's nothing I can do about it? Or is there something I
can do about it? Any suggestions? (other than "don't run SETI"; I
already kill when I notice how slow my first run of the day is.)
Thanks in advance.
Phil Wasson
Managing Editor Inc.
_______________________________________________
macnetworkprog mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/macnetworkprog
Do not post admin requests to the list. They will be ignored.