From owner-freebsd-cluster Thu Dec 16 7:46:16 1999 Delivered-To: freebsd-cluster@freebsd.org Received: from ns1.yes.no (ns1.yes.no [195.204.136.10]) by hub.freebsd.org (Postfix) with ESMTP id 78BF514C39 for ; Thu, 16 Dec 1999 07:46:12 -0800 (PST) (envelope-from eivind@bitbox.follo.net) Received: from bitbox.follo.net (bitbox.follo.net [195.204.143.218]) by ns1.yes.no (8.9.3/8.9.3) with ESMTP id QAA16001; Thu, 16 Dec 1999 16:46:09 +0100 (CET) Received: (from eivind@localhost) by bitbox.follo.net (8.8.8/8.8.6) id QAA78295; Thu, 16 Dec 1999 16:46:07 +0100 (MET) Date: Thu, 16 Dec 1999 16:46:07 +0100 From: Eivind Eklund To: Jeroen Ruigrok van der Werven Cc: cluster@FreeBSD.ORG Subject: Re: Project homepage and definitions Message-ID: <19991216164607.H62815@bitbox.follo.net> References: <19991216155348.C68446@lucifer.bart.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <19991216155348.C68446@lucifer.bart.nl>; from asmodai@bart.nl on Thu, Dec 16, 1999 at 03:53:48PM +0100 Sender: owner-freebsd-cluster@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Thu, Dec 16, 1999 at 03:53:48PM +0100, Jeroen Ruigrok van der Werven wrote: > Added a few more references. (still sifting through Eivind's list.) > > I am also in the prospect in writing a text on what exactly high > availability and clustering is. > > basically, at least what I know is that: > > high availability is a more sophisticated form of load balancing High Availability is used somewhat fuzzily, but represent 'Making a system be available more of the time than usual'. This might be 24/7 (e.g. web servers), or it might be *always* during the 8/5 working hours, with "anything goes" during the rest of the time (e.g, stock exchange systems). One common definition of a High Availabilable System is 'Anything without any SPOF (Single Point Of Failure)', differing from 'Classic Fault Tolerant' (which is defined as the usual tell-me-tree-times or similar hardware level fault tolerance). > clustering is basically the grouping of stations and the sharing of > processes between those stations. Also parallization plays an important > aspect. Clustering is also fuzzy, but do not need to include process migration (and actually usually does not). The definition I find myself most comfortable with is 'Any set of computers with an SSI (Single System Image) seen from some point of view'. Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-cluster" in the body of the message