From owner-freebsd-current Wed Apr 29 14:15:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA01074 for freebsd-current-outgoing; Wed, 29 Apr 1998 14:15:03 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from friley585.res.iastate.edu (friley585.res.iastate.edu [129.186.167.85]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA01044 for ; Wed, 29 Apr 1998 14:14:55 -0700 (PDT) (envelope-from ccsanady@friley585.res.iastate.edu) Received: from friley585.res.iastate.edu (loopback [127.0.0.1]) by friley585.res.iastate.edu (8.8.8/8.8.8) with ESMTP id QAA00921 for ; Wed, 29 Apr 1998 16:14:51 -0500 (CDT) (envelope-from ccsanady@friley585.res.iastate.edu) Message-Id: <199804292114.QAA00921@friley585.res.iastate.edu> X-Mailer: exmh version 2.0.1 12/23/97 cc: current@FreeBSD.ORG Subject: freebsd-net mailing list? (Re: Bandwidth throttling etc.) In-reply-to: Your message of "Fri, 24 Apr 1998 19:16:36 CDT." <199804250016.TAA01428@friley585.res.iastate.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 29 Apr 1998 16:14:51 -0500 From: Chris Csanady Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG There seems to be a growing interest in networking--would it be possible to have a freebsd-net mailing list created for technical discussion of networking? Aside from that, I have to wonder if the below part of my earlier message was simply missed, or just plain ignored. I am describing a serious and real problem here--something that needs to be dealt with. It can cause *horrible* performance for some applications. Earlier someone posted about a web server getting 5 transactions/sec. This also had terrible effects on our clustering performance until we disabled the delayed acks. The problem basically causes throughput to effectively drop to *zero* for some message sizes. That being said, my intent is not whine and complain about this until someone fixes it. Rather, I would simply like to see some discussion of the problems.. Chris Csanady -- [ talk of increasing cluster size deleted ] >Whether we use mbufs clusters or whatever, the packets should be >contiguous. I think that this in itself causes somewhat of a problem >with regards to the current socket buffer handling. Currently, the fact >that the advertised window is calculated based on the total memory consumed >by the mbufs--and not only the data, is causing serious performance >problems. This interaction with delayed acks is the only Really Bad thing >wrong with our stack that I have noticed. > >I believe that BSDI has done some work to deal with this by creating an >m_compress function that collapses wasted space. [ other random networking stuff deleted ] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message