From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on dcvr.yhbt.net X-Spam-Level: X-Spam-ASN: AS6939 64.71.128.0/18 X-Spam-Status: No, score=-2.0 required=3.0 tests=AWL,BAYES_00, MSGID_FROM_MTA_HEADER shortcircuit=no autolearn=unavailable version=3.3.2 Path: news.gmane.org!not-for-mail From: Eric Wong Newsgroups: gmane.comp.lang.ruby.kgio.general Subject: Re: [PATCH 3/3] add `#kgio_writev` and `#kgio_trywritev` Date: Thu, 31 May 2012 14:56:02 -0700 Message-ID: <20120531215602.GA27669@dcvr.yhbt.net> References: <1338386216-14568-1-git-send-email-funny.falcon@gmail.com> <1338386216-14568-3-git-send-email-funny.falcon@gmail.com> <20120530203915.GB17661@dcvr.yhbt.net> <20120531211404.GA24592@dcvr.yhbt.net> NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Trace: dough.gmane.org 1338501377 8840 80.91.229.3 (31 May 2012 21:56:17 GMT) X-Complaints-To: usenet@dough.gmane.org NNTP-Posting-Date: Thu, 31 May 2012 21:56:17 +0000 (UTC) To: kgio@librelist.com Original-X-From: kgio@librelist.com Thu May 31 23:56:16 2012 Return-path: Envelope-to: gclrkg-kgio@m.gmane.org List-Archive: List-Help: List-Id: List-Post: List-Subscribe: List-Unsubscribe: Precedence: list Original-Sender: kgio@librelist.com Xref: news.gmane.org gmane.comp.lang.ruby.kgio.general:170 Archived-At: Received: from zedshaw.xen.prgmr.com ([64.71.167.205]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1SaDLr-0003C8-EX for gclrkg-kgio@m.gmane.org; Thu, 31 May 2012 23:56:15 +0200 Received: from zedshaw.xen.prgmr.com (localhost [IPv6:::1]) by zedshaw.xen.prgmr.com (Postfix) with ESMTP id DE87221DD1C for ; Thu, 31 May 2012 22:04:09 +0000 (UTC) Eric Wong wrote: > True about kgio for mainly for sockets, but 1M might be too small for > long fat networks (high latency, high bandwidth). > > /proc/sys/net/ipv4/tcp_wmem on my stock Linux 3.4 system says the > write buffer can grow up to 4M. I've checked some setups I use (75-120ms over 1 gigabit) and those have the max tcp_wmem set to 16M(!). I've also known (and pity) folks that have to deal with ~500ms latency on their LFNs.