From owner-freebsd-hackers Mon May 20 18:16:33 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id SAA08218 for hackers-outgoing; Mon, 20 May 1996 18:16:33 -0700 (PDT) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id SAA08212 for ; Mon, 20 May 1996 18:16:30 -0700 (PDT) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.7.5/CET-v2.1) with SMTP id KAA04874; Tue, 21 May 1996 10:16:19 +0900 (JST) Date: Tue, 21 May 1996 10:16:18 +0900 (JST) From: Michael Hancock To: Terry Lambert cc: bsdi-users@bsdi.com, freebsd-hackers@freebsd.com Subject: Re: ISDN Compression Load on CPU In-Reply-To: <199605202119.OAA28500@phaeton.artisoft.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Mon, 20 May 1996, Terry Lambert wrote: > > SDLComm has a new PCI card called the RISCom/Pri. Primary rate has 23 > > 64Kbps B-Channels. I wonder how many cards can be put into a P90 box > > running BSD Unix before the compression overloads the CPU? > > The RISC CPU you mean.. dunno. Probably what I should've meant. I'm not sure how the processing would be divided between the card and the host CPU. > > How heavy is the computational load of something like STAC compression? > > As a percentage of CPU: depends on your CPU. > > As a percentage of file I/O overhead: ~17%, assuming average cache > locality, degrading to ~36% for cache-busting (like IOZone). This > was on a 486 DX/2-66, so your mileage may vary. > -mh