From owner-freebsd-hackers Thu May 23 12:51:46 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA13020 for hackers-outgoing; Thu, 23 May 1996 12:51:46 -0700 (PDT) Received: from kitten.mcs.com (Kitten.mcs.com [192.160.127.90]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id MAA13015 for ; Thu, 23 May 1996 12:51:44 -0700 (PDT) Received: from venus.mcs.com (root@Venus.mcs.com [192.160.127.92]) by kitten.mcs.com (8.7.5/8.6.9) with SMTP id OAA12657; Thu, 23 May 1996 14:51:41 -0500 (CDT) Received: by venus.mcs.com (/\==/\ Smail3.1.28.1 #28.5) id ; Thu, 23 May 96 14:51 CDT Message-Id: Subject: Re: ISDN Compression Load on CPU To: dennis@etinc.com (Dennis) Date: Thu, 23 May 1996 14:51:40 -0500 (CDT) From: "Karl Denninger, MCSNet" Cc: hackers@freebsd.org In-Reply-To: <199605231934.PAA29049@etinc.com> from "Dennis" at May 23, 96 03:34:55 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Karl D. writes... > > >Again, note - no spinning media, no PC-bus issues.... > > I really hate this garbage argument. Novell servers with uptimes > over a year are commonplace...PC bus, spinning media and all. > you only have problems with drives that are too fast, too hot and > too overworked, which simply isnt the case with a router scenario. > > Dennis Yeah, and I have BSD boxes with uptimes in the hundreds of days too. I also have CISCOs under management with uptimes measured in *years*. One in particular with an uptime of over *four* years. The *normal* reason that these CISCOs reboot is because the power goes away (usually intentionally and for maintenance purposes). An awful lot of the *CORE* hardware on my backbone has never had an unsolicited reload, and it works VERY hard. In general, moving parts mean lower reliability. Further, not one PC (or component) builder in 1000 knows item #1 about thermal engineering, which is why all those drives, power supplies, and motherboards burn up and burn out. Fans on processors? With REAL MTBFs in the single-digit-thousand hour ranges? Yeah, right. I'll cook my eggs on the stove rather than on my Pentium, thank you very much. There's nothing wrong with a PC architecture machine per-se. Its the people who build them that are all air between the ears. This includes, by the way, some awfully big names in the industry. The problem is money. If you do the thermal engineering studies, and properly design airflow and fan arrays for a PC, and work around and expect for one of these to be "failed" (and still not burn up) you just put enough on the price of the average PC that its no longer "cheaper" than the dedicated hardware built specifically for the job. Further, there is absolutely no contest when it comes to density in the dial access marketplace. ASCEND hardware wins hands-down over any and all other comers in this market at present, period. And they are quite (but not perfectly) clueful in both the ISDN and analog arenas. That's why National ISPs, and smaller regional ISPs who know what they're doing, are choosing them over the other contenders -- including the so-called "pc router in a box" folks. The "PC card router" people have a niche, yes. But that equipment is just not in the same class as the mainstream dedicated hardware built specifically for the job at hand. -- -- Karl Denninger (karl@MCS.Net)| MCSNet - The Finest Internet Connectivity Modem: [+1 312 248-0900] | T1 from $600 monthly; speeds to DS-3 available Voice: [+1 312 803-MCS1] | 21 Chicagoland POPs, ISDN, 28.8, much more Fax: [+1 312 248-9865] | Email to "info@mcs.net" WWW: http://www.mcs.net/ ISDN - Get it here TODAY! | Home of Chicago's only FULL Clarinet feed!