Date: Sat, 11 Dec 2021 15:21:16 +0300 From: Mehmet Erol Sanliturk <m.e.sanliturk@gmail.com> To: Sami Halabi <sodynet1@gmail.com> Cc: "beepc.ch" <xpetrl@beepc.ch>, FreeBSD Current <freebsd-current@freebsd.org> Subject: Re: Benchmarks: FreeBSD 13 vs. NetBSD 9.2 vs. OpenBSD 7 vs. DragonFlyBSD 6 vs. Linux Message-ID: <CAOgwaMsucChUxRx9QzDkx3AR_N49uey_ssHnEZNod5dAMRpQvw@mail.gmail.com> In-Reply-To: <CAEW%2BogaXmG6hm%2B9NcKYfYDJB7mx2V=cN8DoyvZvUwMhkvkJNvQ@mail.gmail.com> References: <CA%2BGLnbgVGghYAYPbQfu0H0cGvXxk-v0jAZTxLLz%2BhRn5eXjP0g@mail.gmail.com> <f8e569a4-3510-5a91-62b2-f9080b197ebc@beepc.ch> <CAEW%2BogaXmG6hm%2B9NcKYfYDJB7mx2V=cN8DoyvZvUwMhkvkJNvQ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--00000000000077f7b105d2dde2bd Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable I can say that these "micro" benchmarks are not "so much" useful . When such comparisons are made based on "abstract" views , they DO NOT SHOW very much utility . Over previous years I am always stressing that proper comparisons would be based on a "specified" workload : Which distributions is more suitable for that "specified" work load ? Testing / benchmarking should be performed with respect to such criteria , for example , continuously "a day work necessary for a profession" , "a web server for a 'specified' task" , "a NFS server working for a 'specified' application" . Then comparisons of parts causing significant differences may generate useful improvement possibilities . When we see *BSD distributions as a single group , it is obvious that everyone has its own priorities to realize . The same is true also for Linux distributions . Without taking such differences into consideration , reaching some conclusions about them would only be a waste of time . With my best regards , Mehmet Erol Sanliturk On Sat, Dec 11, 2021 at 1:40 PM Sami Halabi <sodynet1@gmail.com> wrote: > Hi, > I see these claims over and over. > So I must ask. > Is there any tunibg guide(s) to make the default not conservative in a > regrding to several use cases like using as web server? How to Utilize gp= u > maybe? > I know there are few network (aka routing / forwarding) guides.. but mayb= e > instead of that superior feeling "oh they are linuxish and knoe shit" may= be > better supply the tuning needed to get better results? > And I'm not talking to get an engineer to analyze the tests case.. > Maybe the linux defaults fit better for most use cases rather than being > conservative?? > > Just to be clear I almost not used linux and always freebsd for simplicit= y > usage.. but I must say it makes me wonder > > Sami > > =D7=91=D7=AA=D7=90=D7=A8=D7=99=D7=9A =D7=A9=D7=91=D7=AA, 11 =D7=91=D7=93= =D7=A6=D7=9E=D7=B3 2021, 11:52, =D7=9E=D7=90=D7=AA beepc.ch =E2=80=8F<xpetr= l@beepc.ch>: > > > > I am surprised to see that the BSD cluster today has much worse > > performance > > > than Linux. > > > What do you think of this? > > > > "Default" FreeBSD install setting are quite conservative. > > The Linux common distros are high tuned, those benchmark is in my > > opinion comparison of apples and oranges. > > > > Comparing "default" FreeBSD install with "default" Slackware install > > would be more interesting, because Slackware builds are at most vanilla= . > > > > > --00000000000077f7b105d2dde2bd--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAOgwaMsucChUxRx9QzDkx3AR_N49uey_ssHnEZNod5dAMRpQvw>