Date: Fri, 31 Jul 2015 09:31:53 +0200 From: Willem Jan Withagen <wjw@digiware.nl> To: Erich Dollansky <erichsfreebsdlist@alogt.com> Cc: Josh Paetzel <jpaetzel@FreeBSD.org>, freebsd-hardware@freebsd.org Subject: Re: L2 cache errors??? Message-ID: <55BB2469.5040507@digiware.nl> In-Reply-To: <20150731132251.5b8581c2@X220.alogt.com> References: <55B7B8FA.2060800@digiware.nl> <55B7C059.5020701@sentex.net> <55B7CCA1.4020906@digiware.nl> <55B7D24B.5060709@FreeBSD.org> <55B7DBBF.2090009@digiware.nl> <20150731132251.5b8581c2@X220.alogt.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 31/07/2015 07:22, Erich Dollansky wrote: > Hi, > > On Tue, 28 Jul 2015 21:45:03 +0200 > Willem Jan Withagen <wjw@digiware.nl> wrote: > >> On 28/07/2015 21:04, Josh Paetzel wrote: >> >> Offlining CPus, cool. > > and bringing them back online when the problem is fixed. The hardware > there supports that things get changed while the system is running. A > PC costs normally less than the extra hardware required to do this. Yes, I can imagine things being expensive. Probably like high-end routers... There you can swap also just about anything.. Expensive got a complete new meaning when I saw a set of 3 core routers being delivered at a friends ISP with a pricetag > 1.000.000 euros... Fortunately it was list price, but even still: A lot of money. But then still you'd swap a processor board, running on the spare. And I guess there we offline the whole board. Last time I heard about things like this in computers, we were talking IBM mainframes, or Tandem. Both long time ago. Obviously haven't done much in High availability lately. Although I've grown to look at HA as: don't put it all in one (expensive) box, but get more (cheaper) boxes, But I guess there are places where this doesn't work. --WjW
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?55BB2469.5040507>