Re: How fast is your server?

This WebDNA talk-list message is from

2002

It keeps the original formatting. numero = 41866
interpreted = N
texte = On 7/23/02 2:14 PM, Andrew Simpson mashed the following keys :> Its a fair enough comment that the two platforms may not account for a> 'tick' in the same way but the fact remains that the PC is visibly faster> than the Mac. Its something that can be measured by counting in your head...Yeah, but again, the PC is 3.4x the clockspeed of the Mac and completed thetask 2.6x as fast. (if the ticks were even)The 533DP's strength is the fact that it has dual procs, which this testharldy accounts for...I would certainly expect this result, but when you have 30 simultaniousconnections hitting apache, webcatalog and whatever else (communigate?) I would expect the difference to be lower still.The debate rages on :)I would love to see a 1gig xserve and a 1 or 2 ghz pentium do the test.Especially a dual, dual.In a week or so we should have our xserve, so we have that long to devise amore betterer, thorough-ish test, that displays the result in seconds... Isecond is still more or less a second on a pc, isn't it? > > I love mac's but I also love money and seen as how here in new zealand you> can buy 6 PC's for the price of an Xserve, the Xserve is already at a big> disadvantage.> > I would be interested in a proper test also as Aaron has sugested.> > I will try and device some better coding challenges to test later on> tonight...> > Unless smith micro has something it can share?> > On 24/7/02 1:27 AM, John Peacock wrote:> >> Andrew Simpson wrote:>>> >>> The PC completed this task in 243 ticks while the mac took 637 ticks.>> >> Ticks on PC != Ticks on Mac>> >> The underlying operating system measures time in different ways; I believe on>> the Mac, a tick is 1/8 of a second, whereas on a PC it is 100 milliseconds>> (0.1 >> second). So you are not measuring the same thing.>> >> That being said, the disk processing time of writing 20000 times to a file is>> likely going to swamp any other variable (milliseconds vs nanoseconds). It>> will >> vary by O/S (classic Mac is not tuned as a server O/S), RAM, disk subsystem>> (caching controller), and disk geometry itself. I don't think this test can>> measure anything useful, unless you use the same box and vary some of the>> parameters (add RAM, use a caching controller, use a 10k disk drive).>> >> John-------------------------------------------------------------This message is sent to you because you are subscribed to the mailing list .To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://search.smithmicro.com/ Associated Messages, from the most recent to the oldest:

Re: How fast is your server? (Alain Russell 2002)
Re: How fast is your server? (Alain Russell 2002)
Re: How fast is your server? (Bob Minor 2002)
Re: How fast is your server? (Andrew Simpson 2002)
Re: How fast is your server? (Kenneth Grome 2002)
Re: How fast is your server? (Andrew Simpson 2002)
Re: How fast is your server? (Aaron Lynch 2002)
Re: How fast is your server? (Andrew Simpson 2002)
Re: How fast is your server? (John Peacock 2002)
Re: How fast is your server? (Alain Russell 2002)
Re: How fast is your server? (Aaron Lynch 2002)
Re: How fast is your server? (Aaron Lynch 2002)
Re: How fast is your server? (Andrew Simpson 2002)
Re: How fast is your server? (Aaron Lynch 2002)
How fast is your server? (Andrew Simpson 2002)

On 7/23/02 2:14 PM, Andrew Simpson mashed the following keys :> Its a fair enough comment that the two platforms may not account for a> 'tick' in the same way but the fact remains that the PC is visibly faster> than the Mac. Its something that can be measured by counting in your head...Yeah, but again, the PC is 3.4x the clockspeed of the Mac and completed thetask 2.6x as fast. (if the ticks were even)The 533DP's strength is the fact that it has dual procs, which this testharldy accounts for...I would certainly expect this result, but when you have 30 simultaniousconnections hitting apache, webcatalog and whatever else (communigate?) I would expect the difference to be lower still.The debate rages on :)I would love to see a 1gig xserve and a 1 or 2 ghz pentium do the test.Especially a dual, dual.In a week or so we should have our xserve, so we have that long to devise amore betterer, thorough-ish test, that displays the result in seconds... Isecond is still more or less a second on a pc, isn't it? > > I love mac's but I also love money and seen as how here in new zealand you> can buy 6 PC's for the price of an Xserve, the Xserve is already at a big> disadvantage.> > I would be interested in a proper test also as Aaron has sugested.> > I will try and device some better coding challenges to test later on> tonight...> > Unless smith micro has something it can share?> > On 24/7/02 1:27 AM, John Peacock wrote:> >> Andrew Simpson wrote:>>> >>> The PC completed this task in 243 ticks while the mac took 637 ticks.>> >> Ticks on PC != Ticks on Mac>> >> The underlying operating system measures time in different ways; I believe on>> the Mac, a tick is 1/8 of a second, whereas on a PC it is 100 milliseconds>> (0.1 >> second). So you are not measuring the same thing.>> >> That being said, the disk processing time of writing 20000 times to a file is>> likely going to swamp any other variable (milliseconds vs nanoseconds). It>> will >> vary by O/S (classic Mac is not tuned as a server O/S), RAM, disk subsystem>> (caching controller), and disk geometry itself. I don't think this test can>> measure anything useful, unless you use the same box and vary some of the>> parameters (add RAM, use a caching controller, use a 10k disk drive).>> >> John-------------------------------------------------------------This message is sent to you because you are subscribed to the mailing list .To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://search.smithmicro.com/ Aaron Lynch

DOWNLOAD WEBDNA NOW!

Re: How fast is your server?

2002

Top Articles:

Related Readings: