From owner-freebsd-virtualization@FreeBSD.ORG Sat Jul 11 20:05:06 2009 Return-Path: Delivered-To: freebsd-virtualization@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 00B07106566B for ; Sat, 11 Jul 2009 20:05:06 +0000 (UTC) (envelope-from toreason@fastmail.fm) Received: from out1.smtp.messagingengine.com (out1.smtp.messagingengine.com [66.111.4.25]) by mx1.freebsd.org (Postfix) with ESMTP id CB6388FC15 for ; Sat, 11 Jul 2009 20:05:05 +0000 (UTC) (envelope-from toreason@fastmail.fm) Received: from compute2.internal (compute2.internal [10.202.2.42]) by out1.messagingengine.com (Postfix) with ESMTP id D96933B042A for ; Sat, 11 Jul 2009 16:05:04 -0400 (EDT) Received: from web7.messagingengine.com ([10.202.2.216]) by compute2.internal (MEProxy); Sat, 11 Jul 2009 16:05:04 -0400 Received: by web7.messagingengine.com (Postfix, from userid 99) id A12961ADD1; Sat, 11 Jul 2009 16:05:04 -0400 (EDT) Message-Id: <1247342704.20289.1324595713@webmail.messagingengine.com> X-Sasl-Enc: iUMmgc8U+6l4ADM01z3FLnKchFD2uGXyq6r2UtOkGd5t 1247342704 From: "V S P" To: undisclosed-recipients:; Cc: freebsd-virtualization@freebsd.org Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 X-Mailer: MessagingEngine.com Webmail Interface In-Reply-To: <4A58EEF5.4020502@elischer.org> References: <1247300972.7362.1324546425@webmail.messagingengine.com> <4A58EEF5.4020502@elischer.org> Date: Sat, 11 Jul 2009 16:05:04 -0400 Subject: Re: single image OS with multiple freebsd machines X-BeenThere: freebsd-virtualization@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "Discussion of various virtualization techniques FreeBSD supports." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 11 Jul 2009 20:05:06 -0000 hi, I agree, and hoping that something like that is already out there especially given freebsd separation for virtual spaces/cpus for jail support it just the other way around adding resources 'back' into the pool After learning about kerrighed I am thinking that various hadoop/ map reduce systems and clustered database servers (when they are clustered for performance and not fault-tolerance) -- are really incorrect approaches for multi-machine parallelism --instead having it done by OS resources is really the right approach as it it basically tells application developer: " as long as you design your system to run concurrently on a single machine our OS will automatically scale it across more than one machine " this is very powerful -- and I was hoping something like this is also available in the BSD system. On Sat, 11 Jul 2009 08:58 -0700, "Julian Elischer" wrote: > V S P wrote: > > Hi, > > new to the list > > > > came across of > > http://www.kerrighed.org/wiki/index.php/Main_Page > > > > Which a linux-based system that virtualizes individual machines > > into one cohesive OS so 4 machines with 4GB and 2 cpus each > > look to applications as one machine with 16GB and 8CPUs > > certainly looks interesting. > I believe it is no more nor less feasible on FreeBSD than Linux. -- Vlad P author of C++ ORM http://github.com/vladp/CppOrm/tree/master -- http://www.fastmail.fm - Send your email first class