Re: Benchmark (Phoronix): FreeBSD 9.0-RC2 vs. Oracle Linux 6.1 Server

From: Igor Mozolevsky <igor_at_hybrid-lab.co.uk>
Date: Wed, 21 Dec 2011 22:33:24 +0000
On 21 December 2011 22:03, Freddie Cash <fjwcash_at_gmail.com> wrote:
> On Wed, Dec 21, 2011 at 1:49 PM, Johan Hendriks <joh.hendriks_at_gmail.com> wrote:
>> Nice page, but one thing i do not get is the following.
>>
>> [quote]
>> If you compare FreeBSD / GCC 4.2.1 against, for example, Ubuntu / GCC 4.7
>> then the results are unlikely to tell you anything meaningful about FreeBSD
>> vs Ubuntu.
>> [/quote]
>>
>> That is a little strange in my opinion.
>> It tells me that FreeBSD falls more and more behind on Linux.
>> The reason is or could be that FreeBSD cannot or will not include GCC 4.7
>> and that FreeBSD will not be on par with Linux anymore.
>
> When benchmarking two systems, you need to make sure that everything
> possible is the same (constants) and that the only differences between
> the two systems are what you want to benchmark (variables).

Yes and no, but to be perfectly frank, the statement, as it stands, is
a bit of a nonsense. Let me illustrate in a different way. This is
macro~ vs micro~comparison of systems and depends on what you are
trying to get out of the benchmark. Using the same argument one can
say that Ferrari F430 vs Toyota Prius is a meaningless comparison
because the under-the-hood equipment is different.

Now, it is absolutely correct to say that in A vs B comparisons, only
one thing should be changed and the rest should remain constant. The
important thing is, however, to determine the scope of your benchmark:
you are not benchmarking a component of A vs a component of B, but you
are benchmarking A as a whole system and B as a whole system. Thus,
the thing that changes is the system itself. Going back to F430 vs
Prius, you first decide what you want to benchmark (acceleration, top
speed, fuel consumption, ride comfort, &c) then you measure that
aspect in each of the system---you are not looking at the wiring,
engine, wheels, &c individually but *at a whole system*. You use the
same route, time of day, driver, drive pattern, weather conditions,
&c, the only thing that changes is the car! Similarly, FreeBSD vs
Linux, you want to a) determine what metric you want to benchmark (NFS
throughput, HTTP client handling, SMPT throughput, prime number
computation) and b) *scientifically* measure the system against that
metric... This would essentially amount to having identical set up and
tests, and only changing the hard disks (one containing Linux and
another one containing FreeBSD). I don't see why this is such a
difficult concept to grasp.


Cheers,

--
Igor M. :-)
Received on Wed Dec 21 2011 - 21:34:07 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:22 UTC