We've tried to increase the pipe buffer (not bufmod) on FreeBSD this week. Besides some tricky implementation details the performance gains in a pure 'copy data' benchmark are *astonishing*. Copy time for a 5140 buffer (from 512) goes down from 4 mins to 2 mins and for a 20480 buffer it goes down to 74 seconds (not a linear win but still something to consider). We'll post patches for FreeBSD next month.
Chris On Fri, Jul 10, 2009 at 8:56 PM, rickey c weisner <[email protected]> wrote: > You can push a bufmod on the fifo and use messages up to the maximum > streams message size. > rick > > On Fri, Jul 10, 2009 at 10:48:29AM -0500, Bob Friesenhahn wrote: >> X-Authority-Analysis: v=1.0 c=1 a=CJn96HSh7NQA:10 >> a=lzmFrgBnj+XnLHvemvnKkQ==:17 a=RLQEIbXJAAAA:8 a=tx7JrHvRAAAA:8 >> a=ep_KMAzDAAAA:8 a=b2JNCujJsANkwUJLj04A:9 a=uVteT0YrbvIAIDKuawkA:7 >> a=7_-5LuXN-4LWqWdEruy4W3aRDXUA:4 a=TC1Ce7_l9HwA:10 >> X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on >> oss-mail1.opensolaris.org >> X-Spam-Level: >> X-Spam-Status: No, score=-2.6 required=5.0 tests=AWL,BAYES_00 >> autolearn=unavailable version=3.2.5 >> X-Original-To: [email protected] >> Delivered-To: [email protected] >> Date: Fri, 10 Jul 2009 10:48:29 -0500 (CDT) >> From: Bob Friesenhahn <[email protected]> >> X-X-Sender: [email protected] >> To: Roland Mainz <[email protected]> >> In-Reply-To: <[email protected]> >> X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.2.2 >> (blade.simplesystems.org [65.66.246.90]); >> Fri, 10 Jul 2009 10:48:30 -0500 (CDT) >> Cc: OpenSolaris Performance/HPC discussions <[email protected]>, >> OpenSolaris Shell discussions <[email protected]>, >> Garrett D'Amore <[email protected]> >> Subject: Re: [perf-discuss] Changing the default buffer sizes for pipes ? >> X-BeenThere: [email protected] >> X-Mailman-Version: 2.1.12 >> Precedence: list >> List-Id: Performance General Discussion <perf-discuss.opensolaris.org> >> List-Unsubscribe: <http://mail.opensolaris.org/mailman/options/perf-discuss>, >> <mailto:[email protected]?subject=unsubscribe> >> List-Archive: <http://mail.opensolaris.org/pipermail/perf-discuss> >> List-Post: <mailto:[email protected]> >> List-Help: <mailto:[email protected]?subject=help> >> List-Subscribe: <http://mail.opensolaris.org/mailman/listinfo/perf-discuss>, >> <mailto:[email protected]?subject=subscribe> >> Errors-To: [email protected] >> >> On Fri, 10 Jul 2009, Roland Mainz wrote: >>>> >>>> Some applications may misbehave or lock-up if the size of the pipe >>>> buffer is changed. >>> >>> Erm... why ? >> >> You have already noticed that the size is hard-coded in Solaris >> applications (by PIPE_BUF) since the dawn of time. Pipes offer >> properties such as atomic reads and writes. PIPE_BUF declares the >> maximum size of an atomic write. Some applications depend on this FIFO >> behavior for message passing. >> >> You should feel happy that Solaris is using 5120 rather than 512 like >> FreeBSD. >> >> Bob >> -- >> Bob Friesenhahn >> [email protected], http://www.simplesystems.org/users/bfriesen/ >> GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ >> _______________________________________________ >> perf-discuss mailing list >> [email protected] > > -- > > Rickey C. Weisner > Software Development and Performance Specialist > Principal Field Technologist > Systems Quality Office > cell phone: 615-308-1147 > email: [email protected] > _______________________________________________ > perf-discuss mailing list > [email protected] > -- ^---^ (@)v(@) Chris Pickett | / IT consultant ===m==m=== [email protected] _______________________________________________ perf-discuss mailing list [email protected]
