|Brian Stempin on 24 Mar 2008 09:58:38 -0700|
Oh, one other thing:
Are you being careful between MBits and MBytes? There's a pretty large difference. A 100Mbit/sec connection = 12.5 MBytes/sec (remember: there are 8 bits per byte). Once you count in TCP/IP overhead (usually between 15% and 25%, depending on who you ask, jumbo frame settinsg, etc), 9MBytes/sec is a fair throughput. I want to say that the fastest I've ever gotten on a 100Mbit connection was around 10.2MBytes/sec.
Anyways, all of that text to say, "Even if you had faster hard drives, your network throughput wouldn't go up that much anyway, if at all."
On Mon, Mar 24, 2008 at 12:42 PM, Brian Vagnoni <email@example.com> wrote:
From what I've read network speed auto negotiation is a real throughput killer. Interfaces should be manually set to 100mB full duplex if the card and switch can handle it. Also check for accessive broadcast traffic another throughput killer.
___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug