Re: [SILUG]: [LUNI] Bunk???

[Rich <rich@richnut.com>]

I wish I could remember where I read it, I might have the terminology wrong, 
but the gist of it was that the kernel is still single tasking when it comes 
to giving out some important resources, where is NT does it "right".

Either which way the article was filled with misinformation and the people
who configured the box for linux made a lot of BIG mistakes.

-Rich


"Casey Boone"  said:
>
>
>linus torvalds(sp?) also has stated recently that apache doesnt do
>multithreading properly and it could use some work.  also there is a new
>webserver that benches faster than apache out now ... i think its with the
>debian distro.  instead of spawning a new thread for each incomming
>connection (like apache does) it processes it all in one thread (except for
>cgi).  it keeps a table of what connections are open and what info each
>connection needs etc....
>
>this article was extremely biased IMHO.  they should have also tested out a
>"real" unix like maybe freebsd or sun solaris. and perhaps even a test of
>novell intranetware.
>
>just my thoughts on it
>
>
>casey boone
>
>ps .. i thought the linux kernel itself was multithreaded ... its not? (im
>not talking about support for it happening.. i mean the kernel itself
>running in different threads)
>
>----- Original Message -----
>>From: Rich <rich@richnut.com>
>To: Kara Pritchard <kara@lanscape.net>
>Cc: <silug-discuss@silug.org>
>Sent: Wednesday, April 14, 1999 9:55 AM
>Subject: [SILUG]: Re: [SILUG]: [LUNI] Bunk???
>
>
>>
>>
>>
>> Kara Pritchard  said:
>>
>> [FUD deletia]
>> >
>> >Both Linux and Windows NT Server were tuned to perform optimally under
>> >each of the two workloads.  "We started the tests using standard Red Hat
>> >Linux 5.2 but had to update it because it does not support hardware RAID
>> >controllers and SMP at the same time," said Mindcraft's president, Bruce
>> >Weiner.  "Linux definitely takes more time and resources to tune and to
>> >configure than Windows NT Server.  You have to search the Net to find
>> >the
>> >latest kernel and driver versions to get the highest performance and
>> >most
>> >reliable modules. Then when you're done, Linux fails to deliver the same
>> >level of performance as Windows NT Server on enterprise-class servers."
>>
>> I almost fell out of my chair on this one..
>> Downloading aside it must have took a good 3 minutes to build that kernel
>> from source on a quad procesor box. What a time drain.
>>
>> [more FUD]
>>
>> I dont deny the results of this test. The whole thing was built around
>> multithreading which is an open issue in the Linux kernel. Linux cant send
>out
>> to four nics at once, the kernel is NOT multithreaded. NT is.
>>
>> They also specifically chose a box with ample amounts of everything
>("enterprise
>> class") to skirt the issues with NT on lower-end lower priced systems.
>they use
>> a big SMP box to level the playing field.
>>
>> If they are truly looking for the best price performance a cluster of
>Linux
>> machines should also have been investigated (remember you dont have to pay
>> per cpu licenses) I think this would be an excellent followup/rebuttle
>test.
>> The Dell they talk about starts at roughly $18000 for a config similar to
>what
>> was reported. Add the cost of NT server and you're able to buy over 10
>Linux
>> workstations at the same cost. The goal here was to demonstrate the NT
>performs
>> better than Linux on the Dell PowerEdge, not that NT performs better than
>Linux.
>>
>> -Rich
>>
>>
>> --
>> To unsubscribe, send email to majordomo@silug.org with
>> "unsubscribe silug-discuss" in the body.
>>
>
>
>--
>To unsubscribe, send email to majordomo@silug.org with
>"unsubscribe silug-discuss" in the body.

--
To unsubscribe, send email to majordomo@silug.org with
"unsubscribe silug-discuss" in the body.