From owner-freebsd-questions@FreeBSD.ORG Tue Jan 10 19:50:57 2006 Return-Path: X-Original-To: freebsd-questions@freebsd.org Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7C5F416A41F for ; Tue, 10 Jan 2006 19:50:57 +0000 (GMT) (envelope-from wojtek@tensor.3miasto.net) Received: from chylonia.3miasto.net (chylonia.3miasto.net [213.192.74.6]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0867243D45 for ; Tue, 10 Jan 2006 19:50:51 +0000 (GMT) (envelope-from wojtek@tensor.3miasto.net) Received: from chylonia.3miasto.net (localhost [127.0.0.1]) by chylonia.3miasto.net (8.13.4/8.13.4) with ESMTP id k0AJoYpj014474; Tue, 10 Jan 2006 20:50:34 +0100 (CET) (envelope-from wojtek@tensor.3miasto.net) Received: from localhost (wojtek@localhost) by chylonia.3miasto.net (8.13.4/8.13.4/Submit) with ESMTP id k0AJoXoZ014471; Tue, 10 Jan 2006 20:50:34 +0100 (CET) (envelope-from wojtek@tensor.3miasto.net) X-Authentication-Warning: chylonia.3miasto.net: wojtek owned process doing -bs Date: Tue, 10 Jan 2006 20:50:33 +0100 (CET) From: Wojciech Puchar X-X-Sender: wojtek@chylonia.3miasto.net To: "Marc G. Fournier" In-Reply-To: <20060110125050.A48499@ganymede.hub.org> Message-ID: <20060110204341.U13783@chylonia.3miasto.net> References: <20060110125050.A48499@ganymede.hub.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-questions@freebsd.org Subject: Re: Dual Core vs HyperThreading vs Dual CPU X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 10 Jan 2006 19:50:57 -0000 > I'm going to assume that Dual Core is better (can't believe that they took a > step back) ... but, is how does it rate? I know that HyperThreading is > definitely != Dual CPU ... but how close does Dual Core get? > Dual Core = two physical CPUs, possibly sharing L2 cache. HyperThreading = double sets of registers, machine states etc... for single CPU making it look like 2 CPUs. when one "CPU" has to wait for say memory access, it then executes program from other thread. it means it can be faster than single CPU, but of course much slower than 2 CPUs. because this strategy make cache data shared and often switched between two different threads, it can result slower performance. the net effect is positive in Intel's tests, but very little positive-to little negative in practice like FreeBSD servers. Many users end in running single-CPU kernel because it actually works faster.