Re: Unices are created equal, but ...

JULIAN Elischer (julian@ref.tfs.com)
Sun, 14 Apr 1996 13:55:05 -0700 (PDT)


> ( Did you remember some recent thread in the linux kernel list?)
Sorry I am not on the Linux list..
>
> 2 - People that claim that Unix B is FASTER than Unix A rarely
> indicate the versions of systems (nor the used benchmarks).
> I was expecting that Unix B version 2.0.5 was still a little FASTER that
> Unix A version 1.3.87, and I get the OPPOSITE.
> The difference is that I wrote that my benckmark is "questionnable" and
> give enough informations to guess missing informations.
> Most of Unix end users are not able to guess the MISSING informations.
> They BELEIVE what we CLAIM.

Ok, here is my attempt at the missing information....

I would expect the following to be true between Linix 1.3.x and FreeBSD 2.0.5.
category 1:--Linux faster in:
context switch, including some system calls.
possibly some program startup.
possibly pipe code.
possibly FP emulation.
possibly FP exception handling.
Any test that does a lot of filesystem meta-data manipulation.
e.g. file creation and deletion.

category 2:--FreeBSD 2.0.5 faster in: Anything to do with networking
Anything using a raw tape or disk device.
Any benchmark that loaded the system very heavily,
especially if it produced swapping.
Any benchmark that tested high-speed large sustained
IO to files.

category 3:--Linux and FreeBSD 2.0.5 about equivalent in:
Anything that relies mostly on plain CPU
with no or little OS involvement.
(as both use the same cpu.)

Obvioulsy,
any benchmark that mainly tests category 1 and 3 will get a different result
to one that tests mainly 2 and 3.

My guess is that the Byte Benchmarks don't do much in category 2

>
> 3 - It seems to me that now, current Linux is as FAST as FreeBSD-current.
> Good news!!!!!!!!!!

welll, current Linux is as fast in some places as old FreeBSD..
What you say MIGHT also be true, but this is not proved by your tests.
because:
a/ you only test categories 1 and 3
b/ You do not test current FreeBSD
If you wish to test current FreeBSD and include test from category 2
as well, then I suspect that you might find a different story.

THE REASON:
FreeBSD and Linux developers have done work in different places..
each has strong and weak points.

> It is difficult to have both Linux and FreeBSD in their current version.
> Only linux is up to date on my machine.
> If you are a FreeBSD-current user and if you have about the same
> configuration as mine, can you run the old BYTE benchmark
> and send to me your results?

I don't think it would be useful unless we had EXACTLY the same hardware..
I have seen small differences make up to 50% difference..

Linux made great improvements in the last year.
So did FreeBSD. Most of the areas indicated as being Weak points for FreeBSD
in 2.0.5 were redone in -current. Of course this now means that Other
parts are the "weak points" :)

> Thanks per advance,
>
> Best Regards, Gerard.
>
julian
+----------------------------------+ ______ _ __
| __--_|\ Julian Elischer | \ U \/ / On assignment
| / \ julian@tfs.com +------>x USA \ in a very strange
| ( OZ ) 300 lakeside Dr. oakland CA. \___ ___ | country !
+- X_.---._/ USA+(510) 645-3137(wk) \_/ \\ ><DARWIN>
v LL LL