From: "Darin McBride" Subject: Re: Larry Ellison comments on Microsoft's benchmark Date: 2000/07/07 Message-ID: #1/1 Content-Transfer-Encoding: 7bit References: <39527E0C.E614B483@penev.com> <395554AC.9D413341@penev.com> <8ja49v$1nnk$1@adenine.netfront.net> <3958CCCA.694E57B3@ca.ibm.com> <39616BBF.C3CFA30C@ca.ibm.com> <39625320.12A2DDFA@ca.ibm.com> <2jv85.27607$i5.283945@news1.frmt1.sfba.home.com> <39629B7D.714D7427@ca.ibm.com> <1CA85.28659$i5.293607@news1.frmt1.sfba.home.com> <39634D10.AF8FD314@ca.ibm.com> <39641169.5C934DDC@ca.ibm.com> <39647E61.5C0A238D@ca.ibm.com> <396514F2.C5271109@ca.ibm.com> Content-Type: text/plain; charset="us-ascii" X-Complaints-To: abuse@home.net X-Trace: news1.rdc2.on.home.com 962973480 24.65.66.12 (Fri, 07 Jul 2000 05:38:00 PDT) Organization: @Home Network Canada MIME-Version: 1.0 Reply-To: "Darin McBride" NNTP-Posting-Date: Fri, 07 Jul 2000 05:38:00 PDT Newsgroups: comp.databases.ibm-db2,comp.databases.ms-sqlserver,comp.databases.oracle.server,comp.databases.sybase,microsoft.public.sqlserver.server On Fri, 07 Jul 2000 04:08:47 GMT, Ivana Humpalot wrote: >"Blair Kenneth Adamache" wrote: >> >> You can cross mount disks with shared nothing (also called drive >> mapping). Only one node owns the disk at any time. If the owner node >> fails, the mutual takeover node gets the IP of the failing node, and >> takes over the disk. > >Thank you for the info and the link to IBM documentation. From >what I understand, MSCS clusters are limited to 2 servers per >cluster. So if you have 12 machines you have 6 independent >clusters. Isn't this still 6 times less reliable than Oracle >Parallel Server? Let's use some "real" fake numbers. Let's pretend that each machine has a failure rate of 10%. This is awfully high, but it's a nice, round number. ;-) One machine: 10% failure rate. Twelve machines: 100% - ((100% - 10%)^12) = 71.75% failure rate. 1 2-way mutual-failure takeover cluster: 10% ^ 2 = 1% failure rate. 6 2-way mutual-failure takeover clusters: 100% - ((100% - 1%) ^ 6) = 5.85% failure rate. 12-way mutual-failure takeover cluster: 10% ^ 12 = 10^(-10)% failure rate. "6 times" is a misnomer. But even with a huge overstatement of the failure rates, it can become reasonable fairly quickly. NOTE!!! I did not take reliability engineering courses - only engineering statistics. Which means that while I'm not completely unqualified in reliability analysis, it's close. :-)