Date: Tue, 18 Nov 2008 11:13:40 -0600 From: "Andrew Gould" <andrewlylegould@gmail.com> To: "Jerry McAllister" <jerrymc@msu.edu> Cc: Dan <dan-freebsd-questions@ourbrains.org>, freebsd-questions@freebsd.org Subject: Re: FreeBSD and hardware?? Message-ID: <d356c5630811180913w385ca549l1c76ed246a5ef33e@mail.gmail.com> In-Reply-To: <20081118164947.GF28439@gizmo.acns.msu.edu> References: <7f8a05a80811171609h2f5742cfo2ae003415a99e45f@mail.gmail.com> <20081118121634.S84157@wojtek.tensor.gdynia.pl> <20081118154226.GA20147@ourbrains.org> <20081118165011.G84987@wojtek.tensor.gdynia.pl> <20081118155448.GC20147@ourbrains.org> <20081118164947.GF28439@gizmo.acns.msu.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Nov 18, 2008 at 10:49 AM, Jerry McAllister <jerrymc@msu.edu> wrote: > On Tue, Nov 18, 2008 at 10:54:48AM -0500, Dan wrote: > > > Wojciech Puchar(wojtek@wojtek.tensor.gdynia.pl)@2008.11.18 16:51:16 > +0100: > > >> > > >> Have you used, erm... Linux? Both Linux and FreeBSD run pretty much at > > >> hardware level. You benchmark either, you'll get very close results in > > > > > > for benchmarks doing same thing over and over, or same thing in > parallel > > > linux can even be better. > > > > > > but try running many different tasks in parallel under linux. FreeBSD > > > flies, while linux chokes. > > > > Can you point out some places on the web that confirm this? > > > > I can't point this out between Linux and FreeBSD, but back a few > years ago, when I was involved in benchmarking high performance > systems for purchase here, we found this to often be the case. > Some systems just screamed on certain very parallel tasks, but > practically came to a halt when a mix of tasks were run or even > when trying to edit a script while things were running. Others > were slightly less hot on the highly specialized tasks, but did > well - much better - on the mix. We chose the system that handled > the mix - which ran a BSD UNIX by the way, although a proprietary > version as did most back then. > > Anyway, so, even though I haven't compared FreeBSD and Linux, I am > not surprised to hear someone say there is this sort of difference. > It is possible. Someone might investigate further and put out > some verifiable numbers. > > ////jerry > I don't have verifiable numbers; but I can speak from personal experience. I do complex financial/clinical data analysis for hospitals. I was using MS Access as a front-end. On the server end, I started with Linux and PostgreSQL. I moved from Linux to FreeBSD because during my more complicated series of queries, the Linux system would slow to a crawl. Sometimes, the PostgreSQL server would die. This never happened with FreeBSD. I even added Samba services and a web forum for the department. >From 2000 to 2006, the only unplanned downtime experienced with my PostgreSQL/FreeBSD combo was due to 2 separate, prolonged power outages. When power was restored, the hardware and database servers came back online. Sadly, I no longer work there; and no longer have control over database assets. I read once that: "The difference between the lab and the real world is that, in the lab, there is no difference." I wish I had noted the source. Andrew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d356c5630811180913w385ca549l1c76ed246a5ef33e>