Re: [AOLSERVER] A slight change of subject......
Andrew Piskorski said: Btw, if Sun really does follow through and open-source Solaris like they've said they will, it should be interesting to see what sort of cross-fertilization of ideas and techniques may go on between the Solaris and Linux worlds. Only to close-source it again a year or two later and then blackmail all Linux users for a license? Where have I heard that one before!? ;-) Bas. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
On Wed, Jun 23, 2004 at 10:28:17AM -0500, Elliott Phil Civ AFMSA/SGSID wrote: Just to let you know: The backend in this case is a Sun F6800 w/ 12 cpu, 12 GB RAM running Sol8. When I get a chance I'm thinking of a fronted box with each of the OS platforms hitting the backend. (all of this in my spare time, LOL)--Phil Another DD in his spare time installed debian on a E1, so I think you could do that with a minimal hacking. -- Francesco P. Lovergine -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
Your numbers are very interesting. I've been telling people Sun hardware is just as fast but cheaper and they look at me like I've got 3 eyes or something. v240 2 processor 2 gig ram 2 x 73 gig drives Solaris 9 $6895 1747 c/s Compaq DL 360 2 processor 2 gig ram 2 x 73 gig drives Redhat AS 2.1 $7405 1835 c/s On the low end I run Sun X1's off ebay for $400 or the V100 for $995. Even if they only manage 1% of the v240 numbers that's 17 c/s. More than enough for development and most sites. The Sun is a 64 bit machine and has LOM so you can manage it completely from the serial port. I've never used Compaq. Can you power it up and install it without being there? Not that is matters but did you test the 1 gig or 1.28 gig Sun box. the 1.28 just came out. Also in that config the v210 would be $4945 with the 1 gig processors. Basically $2500 cheaper than the Compaq and within 10% of the performance. Given the variability of benchmarks I'd call it a dead heat in performance and price/performance. It's really more which you'd rather use. I personally think you need a 64bit box to run a database and I'd rather have all Sun than mix Solaris and Linux. I would like to see Xserve numbers. I have run the Xraid and it's very impressive Barry On Jun 23, 2004, at 6:58 PM, Adam Leff wrote: Sure. I think the Ops guys there might still have an Xserve laying around. He's right... we do. I'm still not sure if it's ever been turned on. ;) The bigger question is what the bench marks would be exactly. Bingo. It all depends on the application. Most of the testing that's done on webservers is *usually* how it deals under duress serving static pages... yes, there are dynamic testing benchmarks out there. There are so many flippin' ways we use AOLserver just at AOL that it's completely unfair to say that AOLserver X.X is better on Solaris than it is on Linux. You have to take so many things into consideration (external dependencies, databases, compilation options, OS tuning parameters). Comparing AOLserver as used by AOL.com to AOLserver as used by Moviefone.com or even AOLserver as a backend application layer isn't fair. So that's why we do our best to test each application and its dependencies. Sadly, usually the testing is done after the hardware is purchased. Yay for compressed timeframes. But then again, the prices of x86 hardware (and the associated support contracts) make executives happy. :) That being said, I did a test a little bit ago slamming the begeezus out of an .adp page with a bunch of ns_adp_puts in it (so I was exercising the Tcl interpreters, not just the fastpath stuff) on a few platforms: Sun Fire V240, Solaris 9, AOLserver 3.5.10: 1871 conns/sec Sun Fire V240, Solaris 9, AOLserver 4.0.1: 1747 conns/sec Compaq DL 360, RH AS 2.1, AOLserver 3.5.10: 1880 conns/sec Compaq DL 360, RH AS 2.1, AOLserver 4.0.1: 1835 conns/sec Compaq Proliant, RH AS 3.0, AOLserver 3.5.10: 2220 conns/sec Compaq Proliant, RH AS 3.0, AOLserver 4.0.1: 2256 conns/sec As predicted, Red Hat Advanced Server 3.0 came out on top, most likely due to NPTL. The boxes were all hovering between 60-80% CPU utilization... network saturated. ~Adam Adam Leff AOL Web Operations -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
Barry Books said: v240 2 processor 2 gig ram 2 x 73 gig drives Solaris 9 $6895 1747 c/s Compaq DL 360 2 processor 2 gig ram 2 x 73 gig drives Redhat AS 2.1 $7405 1835 c/s The term DL360 as absolutely useless to identify the hardware in the machine. I bought a truckload of them in 2000 and they were mostly single 933GHz P3s. The current models have 1 or 2 3.2 GHz Xeons, so that comparison isn't fair untill we see what hardware was actualy used! from the serial port. I've never used Compaq. Can you power it up and install it without being there? Newer models come with LOM integrated, not sure how well it works, though. But not as nice as Sun. At an ISP I worked, all we did was put them on the management network, added their MAC to the setup server and told which disk config to use for that MAC, turned it on, waited a while et voila, standard build ready to go. Given the variability of benchmarks I'd call it a dead heat in Not so fast, first see what is really in those boxes! In any case, with Intel, you have the option of going dirt cheap and get redundancy in numbers. This isn't a good strategy if all you need for your app is one fast server and one for fail over. But once you get to the point where you need, say, 10 of them, each of which is allowed to fail at any time, you can just buy yourself some motherboards, CPUs, memory and IDE drives and stick em in a cheap rackmount case. Those are cheap enough to have some spares and you won't have to pay for 24x7x4 support either... Don't get me wrong, I am not anti-Sun at all; I love their boxes. And I know they are faster-per-Hz than their Intel-based counterparts. But I don't believe they are as good value as you say they are! Bas. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
On Jun 24, 2004, at 09:52, Barry Books wrote: The Sun is a 64 bit machine and has LOM so you can manage it completely from the serial port. I've never used Compaq. Can you power it up and install it without being there? The Proliants have a pretty nice management interface that allows you to power-up/down, etc... the console redirection is really bad though. They have something called the management pass-thru that's not too bad, but you can't see the entire boot-up process. The serial on the DL 360s we have is not too bad, but Sun so far is pretty superior in that aspect. We have a team that handles the imaging of our machines before they're passed to us, so we don't really have a need for remote network installs. Not that is matters but did you test the 1 gig or 1.28 gig Sun box. the 1.28 just came out. Also in that config the v210 would be $4945 with the 1 gig processors. Basically $2500 cheaper than the Compaq and within 10% of the performance. I did neglect to mention the box configuration in my first email. My bad. The Sun V240 is a 2x1Ghz UltraSparcIIIi running Sol9... I didn't have a 1.28Ghz box available at the time I was doing the tests. The DL 360 is a 2x2.8Ghz Xeon. The Proliant is a 2x3.2Ghz Xeon. All with 2 GB of RAM. So yes, you could argue that the Proliant results are possible skewed because of the difference in processor speed. That wasn't the question we were trying to answer though... we wanted to find out whether we could get the same performance out of a Linux 2x2 that we currently get out of a Sun 2x2... in the front-end applications that we tested, the answer was yes. That's certainly not ALWAYS the case... but in our findings it was true. I didn't do any scale testing on databases or other backend apps of that nature. ~Adam Adam Leff AOL Web Operations -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
Adam Leff said: The Sun V240 is a 2x1Ghz UltraSparcIIIi running Sol9... I didn't have a 1.28Ghz box available at the time I was doing the tests. The DL 360 is a 2x2.8Ghz Xeon. The Proliant is a 2x3.2Ghz Xeon. All with 2 GB of Wow, that's closer than I thought! Mind you, you can get a 2x 3.2GHz/2GB/2x73GB15K machine for a lot less from Dell than from HP/Compaq. But in the end, I guess when you buy brands, it doesn't seem to matter much in price if you go with Sun or any other. I wonder if these 1GHz Suns also need less juice and stay cooler. Not an unimportant consideration; apperantly Google settled on 1GHz machines for their Ireland data centre for that same reason as they could fit more machines in and thus have more CPU power in total. RAM. So yes, you could argue that the Proliant results are possible skewed because of the difference in processor speed. And cache? That 2.8 probably has 512K, the 3.2 likely 1MB, possibly 2. Bas. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
It supprised me also. I had switched to Intel because of cost but lately I've switch back to Sun. It would be interesting to compare numbers between a Compaq and Dell. You would think they would be similar but I think there is a great deal of difference between a $400 2.8 gig P4 and an $800 2.8 gig P4. I did switch from Dell to Sun, but I never did any detailed beachmarking. My informal testing made me think the Sun was as fast or faster. I've run on Intel (Windows and Linux), OSX and Sun. The great thing about AOLServer is you can switch platforms with very little effort. I've even developed on Sun and deployed on Windows with no problems. I did look up the wattage on the Sun and Compaq and they are similar. I suspect the Sparc chip uses less power but by the time you run the rest of the stuff inside the machine there is not much difference. The X1/V100's are a different story. I think they use around 30 watts, but it probably takes 5 to 10 of them to equal a v240. Whatever you pick take a look at the Apple Xraid. 3.5 terabytes and fibre channel for 10K is a good deal and they work just fine with a v240 and Solaris. I even used Apple's fibre channel card in the Sun box. Rumor has it they also work with Linux. Not only are the cheap per gig I think they beat scsi on performance because for the same price you can get nearly 3x the number of spindles while scsi only has 2x the rotation speed. Barry On Thursday, June 24, 2004, at 02:04PM, Bas Scheffers [EMAIL PROTECTED] wrote: Adam Leff said: The Sun V240 is a 2x1Ghz UltraSparcIIIi running Sol9... I didn't have a 1.28Ghz box available at the time I was doing the tests. The DL 360 is a 2x2.8Ghz Xeon. The Proliant is a 2x3.2Ghz Xeon. All with 2 GB of Wow, that's closer than I thought! Mind you, you can get a 2x 3.2GHz/2GB/2x73GB15K machine for a lot less from Dell than from HP/Compaq. But in the end, I guess when you buy brands, it doesn't seem to matter much in price if you go with Sun or any other. I wonder if these 1GHz Suns also need less juice and stay cooler. Not an unimportant consideration; apperantly Google settled on 1GHz machines for their Ireland data centre for that same reason as they could fit more machines in and thus have more CPU power in total. RAM. So yes, you could argue that the Proliant results are possible skewed because of the difference in processor speed. And cache? That 2.8 probably has 512K, the 3.2 likely 1MB, possibly 2. Bas. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
[AOLSERVER] A slight change of subject......
You guys and gals have way to much fun...that's great! Keep up the good work! Questions: 1) Has anyone noticed what platform (be it Wintel, Linux, Solaris, etc, etc) that AOLSERVER performs better that the others? 2) Has anyone noticed that a particular component on a certain platform performs better or worse than another? I ask these two q's as I've been playing with AOLSvr on several platforms (realizing that the OS's have vast differences in handling load). I'm toying with the front-end being on Wintel with the backend ODB (Illustra, etc) running on a healthy-sized Unix machine. -Phil -Original Message- From: AOLserver Discussion [mailto:[EMAIL PROTECTED] On Behalf Of Dossy Sent: Tuesday, June 22, 2004 4:12 PM To: [EMAIL PROTECTED] Subject: Re: [AOLSERVER] TclX keyed list changes in HEAD On 2004.06.22, Zoran Vasiljevic [EMAIL PROTECTED] wrote: Anyways, the compatibilty layer for following old calls has been done: Tcl_GetKeyedListKeys Tcl_GetKeyedListField Tcl_SetKeyedListField Tcl_DeleteKeyedListField Great, thanks. I'll follow up with folks inside AOL and see if their modules work again. -- Dossy -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
On 2004.06.23, Elliott Phil Civ AFMSA/SGSID [EMAIL PROTECTED] wrote: 1) Has anyone noticed what platform (be it Wintel, Linux, Solaris, etc, etc) that AOLSERVER performs better that the others? I think Win32 performs the worst simply because AOLserver code doesn't exploit the more advanced Win32 features available. Comparing Solaris on SPARC to either Win32 or Linux on x86 isn't really a fair comparison since the architectures are quite different -- a low-end Linux box may well outperform a low-end SPARC box (think: Celeron vs. Ultra2, etc.) but the clock speeds of the CPUs are way different: a 900 MHz Celeron can be had for pennies while a 450 MHz CPU for an Ultra2 is considered fast. A cheap Ultra1 will likely have a 170 MHz CPU, but still probably perform as well as that 900 MHz Celeron x86. Or, maybe not -- I haven't actually done or seen actual benchmarks. Good question ... and I'd love to see some objective measuremens to try and answer them. -- Dossy -- Dossy Shiobara mail: [EMAIL PROTECTED] Panoptic Computer Network web: http://www.panoptic.com/ He realized the fastest way to change is to laugh at your own folly -- then you can let go and quickly move on. (p. 70) -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
Don't forget Mac OS X... ;-) - Nathan Dossy wrote on 6/23/04, 10:32 AM: Good question ... and I'd love to see some objective measuremens to try and answer them. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
On Wed, Jun 23, 2004 at 10:32:32AM -0400, Dossy wrote: Comparing Solaris on SPARC to either Win32 or Linux on x86 isn't really a fair comparison since the architectures are quite different -- a low-end Linux box may well outperform a low-end SPARC box (think: Celeron vs. Ultra2, etc.) but the clock speeds of the CPUs are way different: a 900 MHz Celeron can be had for pennies while a 450 MHz CPU for an Ultra2 is considered fast. A cheap Ultra1 will likely have a 170 MHz CPU, but still probably perform as well as that 900 MHz Celeron x86. Or, maybe not -- I haven't actually done or seen actual benchmarks. Oh well, comparing a Linux/SPARC with a Solaris/SPARC on the same boxi _does_ have sense. I'm quite sure that a Linux with 2.6 kernel should perform better than Solaris. I suspect the same could be also with 2.4, but I'm not so sure, due to multi-threading architecture of aolserver. The same comparison could be done for Linux/x86 vs Solaris/x86. But I'm also quite sure who is the winner. Slowlaris is very slw on x86. -- Francesco P. Lovergine -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
On 2004.06.23, Nathan Folkman [EMAIL PROTECTED] wrote: Don't forget Mac OS X... ;-) PPC Linux vs Mac OS X on the G4/G5 platform? -- Dossy -- Dossy Shiobara mail: [EMAIL PROTECTED] Panoptic Computer Network web: http://www.panoptic.com/ He realized the fastest way to change is to laugh at your own folly -- then you can let go and quickly move on. (p. 70) -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
On Wed, Jun 23, 2004 at 10:32:32AM -0400, Dossy wrote: Comparing Solaris on SPARC to either Win32 or Linux on x86 isn't really a fair comparison since the architectures are quite different -- a low-end Linux box may well outperform a low-end SPARC box (think: Celeron vs. Ultra2, etc.) but the clock speeds of the CPUs are way different: a 900 MHz Celeron can be had for pennies while a 450 MHz CPU for an Ultra2 is considered fast. A cheap Ultra1 will likely have a Practically speaking, I can't imagine why anyone would actually care about those distinctions. The fact of the matter seems to be that for almost everybody, for a given amount of money, x86 hardware is going to give you a whole lot more performance than Sparc hardware. (And that using Solaris rather than Linux doesn't provide any performance advantages, either.) Now, if there are any major exceptions to that, that would be interesting. (At a guess, I rather doubt it, at least when it comes to running AOLserver.) And even then, the only people this would REALLY matter to are folks running large websites with many multiple front-end web server machines. Folks like AOL, in other words. Don't your operations people know? FWIW, several years back, the Bitkeeper guys happened to do some limited apples-to-apples Solaris vs. Linux performance comparisons: Solaris x86 and Linux on identical x86 hardware, and Solaris and Linux on identical Sparc hardware. They cautioned that this was NOT in any way a representative benchmark, is was just some informal tests of how some of their Bitkeeper code performed, but their results were interesting: Linux had a 5x speed advantage over Solaris. At the time, McVoy (formerly a Solaris kernel hacker himself) attributed this as likely due to Solaris locking overhead. All that really proved, of course, was that (at least back then), Solaris CAN be dramatically slower than Linux, under some conditions. And by extension, that operating system, surprisingly to many, CAN sometimes impose very noticeable performance differences. Btw, if Sun really does follow through and open-source Solaris like they've said they will, it should be interesting to see what sort of cross-fertilization of ideas and techniques may go on between the Solaris and Linux worlds. -- Andrew Piskorski [EMAIL PROTECTED] http://www.piskorski.com/ -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
Just to let you know: The backend in this case is a Sun F6800 w/ 12 cpu, 12 GB RAM running Sol8. When I get a chance I'm thinking of a fronted box with each of the OS platforms hitting the backend. (all of this in my spare time, LOL)--Phil -Original Message- From: AOLserver Discussion [mailto:[EMAIL PROTECTED] On Behalf Of Francesco P. Lovergine Sent: Wednesday, June 23, 2004 9:57 AM To: [EMAIL PROTECTED] Subject: Re: [AOLSERVER] A slight change of subject.. On Wed, Jun 23, 2004 at 10:32:32AM -0400, Dossy wrote: Comparing Solaris on SPARC to either Win32 or Linux on x86 isn't really a fair comparison since the architectures are quite different -- a low-end Linux box may well outperform a low-end SPARC box (think: Celeron vs. Ultra2, etc.) but the clock speeds of the CPUs are way different: a 900 MHz Celeron can be had for pennies while a 450 MHz CPU for an Ultra2 is considered fast. A cheap Ultra1 will likely have a 170 MHz CPU, but still probably perform as well as that 900 MHz Celeron x86. Or, maybe not -- I haven't actually done or seen actual benchmarks. Oh well, comparing a Linux/SPARC with a Solaris/SPARC on the same boxi _does_ have sense. I'm quite sure that a Linux with 2.6 kernel should perform better than Solaris. I suspect the same could be also with 2.4, but I'm not so sure, due to multi-threading architecture of aolserver. The same comparison could be done for Linux/x86 vs Solaris/x86. But I'm also quite sure who is the winner. Slowlaris is very slw on x86. -- Francesco P. Lovergine -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
I'm that close to getting it to work on Win 3.11 ;-) Dossy wrote on 23/06/2004, 16:22: On 2004.06.23, Nathan Folkman [EMAIL PROTECTED] wrote: Don't forget Mac OS X... ;-) PPC Linux vs Mac OS X on the G4/G5 platform? -- Dossy -- Dossy Shiobara mail: [EMAIL PROTECTED] Panoptic Computer Network web: http://www.panoptic.com/ He realized the fastest way to change is to laugh at your own folly -- then you can let go and quickly move on. (p. 70) -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank. -- Patrick O'Leary Software Engineer, AOL UK AOL (UK) Ltd. 80 Hammersmith Road, London W14 8UD United Kingdom email: [EMAIL PROTECTED] URL: www.aol.co.uk Tel: +44 (0) 20 7348 8462 Fax: +44 (0) 20 7348 8009 AOL UK's recommended destination for online giving: givenow.org This email, its contents and any files transmitted with it are confidential and may be legally privileged. It is intended solely for the addressee(s) only. If you are not the intended recipient, you must not copy, distribute, or take any action in reliance upon it. If you have received this email in error, please notify us immediately via telephone or fax and delete the material from your computer system. AOL (UK) Ltd is registered in England under number 03462696, with its registered office at 80 Hammersmith Road, London W14 8UD. -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.
Re: [AOLSERVER] A slight change of subject......
Sure. I think the Ops guys there might still have an Xserve laying around. He's right... we do. I'm still not sure if it's ever been turned on. ;) The bigger question is what the bench marks would be exactly. Bingo. It all depends on the application. Most of the testing that's done on webservers is *usually* how it deals under duress serving static pages... yes, there are dynamic testing benchmarks out there. There are so many flippin' ways we use AOLserver just at AOL that it's completely unfair to say that AOLserver X.X is better on Solaris than it is on Linux. You have to take so many things into consideration (external dependencies, databases, compilation options, OS tuning parameters). Comparing AOLserver as used by AOL.com to AOLserver as used by Moviefone.com or even AOLserver as a backend application layer isn't fair. So that's why we do our best to test each application and its dependencies. Sadly, usually the testing is done after the hardware is purchased. Yay for compressed timeframes. But then again, the prices of x86 hardware (and the associated support contracts) make executives happy. :) That being said, I did a test a little bit ago slamming the begeezus out of an .adp page with a bunch of ns_adp_puts in it (so I was exercising the Tcl interpreters, not just the fastpath stuff) on a few platforms: Sun Fire V240, Solaris 9, AOLserver 3.5.10: 1871 conns/sec Sun Fire V240, Solaris 9, AOLserver 4.0.1: 1747 conns/sec Compaq DL 360, RH AS 2.1, AOLserver 3.5.10: 1880 conns/sec Compaq DL 360, RH AS 2.1, AOLserver 4.0.1: 1835 conns/sec Compaq Proliant, RH AS 3.0, AOLserver 3.5.10: 2220 conns/sec Compaq Proliant, RH AS 3.0, AOLserver 4.0.1: 2256 conns/sec As predicted, Red Hat Advanced Server 3.0 came out on top, most likely due to NPTL. The boxes were all hovering between 60-80% CPU utilization... network saturated. ~Adam Adam Leff AOL Web Operations -- AOLserver - http://www.aolserver.com/ To Remove yourself from this list, simply send an email to [EMAIL PROTECTED] with the body of SIGNOFF AOLSERVER in the email message. You can leave the Subject: field of your email blank.