Re: [WebDNA] Server Speed/optomization
This WebDNA talk-list message is from 2009
It keeps the original formatting.
numero = 103642
interpreted = N
texte = On 9/14/09 1:15 PM, "Donovan Brooke"
wrote: > Chris Wilkinson wrote: > [snip] >> Automatically commit databases WAS unchecked... I ran the test several times >> with nothing else running. All fresh installs, all new gear. >> >> Looking at all of the tests, my results don't look right in comparison with >> other set-ups. For instance, a lesser Mini beat me by 100 ticks... WTF? > > > > Actually 77, which is 1.25 seconds. There is also a lesser mini behind you. > > > >> Anecdotally, with the solid state drive installed, this is by far the >> fastest booting and application launching I have ever seen on a computer - >> apps don't even bounce in the dock, they just go... >> >> I will do a more thorough test of the system, but I would like to compare >> with another set-up to pinpoint the bottleneck. The Xbench test results site >> is currently down, so Donovan, would you mind sharing the specs on your >> MacBook Pro, perhaps an Xbench test? And, does .4Ghz more processor make >> that big of a difference for WebDNA performance? Maybe I've got slow RAM? >> >> Any other configuration issues to consider? >> >> Thanks, >> -Chris > > > If you are really into finding that 1.2 second difference to the other > mac mini posted on there.. you'd have to do some more tests of apples to > apples... for > example, you are running OSX Server, where the other appears to be > running OSX client. Yes, not only apache, but other performance > differences could be possible within an OS difference. It also appears > you are running a 3rd Party drive.. which I guess could be a flag since > tests > #5 and #9 are writing to your HD.. What is the bus that you are using?.. > maybe look into your SSD more closely??.. or at least run a speed test > using the normal drives. Also, any RAID going on? What format is the > file system?? etc etc.. (lots of possible variables) > > When all is said and done, like I said, I don't think a 302 is something > to worry too much about for $750 in base hardware. > > ..sorry, no time for testing my own machine right now on Xbench. Maybe > later. > > Donovan > Thanks for all the input... FYI, booting from an external firewire raid 0 took it down to a more acceptable 188 ticks... Not certain if I will run it in production [was going to be for back-up], but WebDNA sure likes to run off raid. Thanks, -Chris
Associated Messages, from the most recent to the oldest:
On 9/14/09 1:15 PM, "Donovan Brooke" wrote: > Chris Wilkinson wrote: > [snip] >> Automatically commit databases WAS unchecked... I ran the test several times >> with nothing else running. All fresh installs, all new gear. >> >> Looking at all of the tests, my results don't look right in comparison with >> other set-ups. For instance, a lesser Mini beat me by 100 ticks... WTF? > > > > Actually 77, which is 1.25 seconds. There is also a lesser mini behind you. > > > >> Anecdotally, with the solid state drive installed, this is by far the >> fastest booting and application launching I have ever seen on a computer - >> apps don't even bounce in the dock, they just go... >> >> I will do a more thorough test of the system, but I would like to compare >> with another set-up to pinpoint the bottleneck. The Xbench test results site >> is currently down, so Donovan, would you mind sharing the specs on your >> MacBook Pro, perhaps an Xbench test? And, does .4Ghz more processor make >> that big of a difference for WebDNA performance? Maybe I've got slow RAM? >> >> Any other configuration issues to consider? >> >> Thanks, >> -Chris > > > If you are really into finding that 1.2 second difference to the other > mac mini posted on there.. you'd have to do some more tests of apples to > apples... for > example, you are running OSX Server, where the other appears to be > running OSX client. Yes, not only apache, but other performance > differences could be possible within an OS difference. It also appears > you are running a 3rd Party drive.. which I guess could be a flag since > tests > #5 and #9 are writing to your HD.. What is the bus that you are using?.. > maybe look into your SSD more closely??.. or at least run a speed test > using the normal drives. Also, any RAID going on? What format is the > file system?? etc etc.. (lots of possible variables) > > When all is said and done, like I said, I don't think a 302 is something > to worry too much about for $750 in base hardware. > > ..sorry, no time for testing my own machine right now on Xbench. Maybe > later. > > Donovan > Thanks for all the input... FYI, booting from an external firewire raid 0 took it down to a more acceptable 188 ticks... Not certain if I will run it in production [was going to be for back-up], but WebDNA sure likes to run off raid. Thanks, -Chris
Chris Wilkinson
DOWNLOAD WEBDNA NOW!
Top Articles:
Talk List
The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...
Related Readings:
WebCat2: Items xx to xx shown, etc. (1997)
unable to launch acgi in WebCat (1997)
Running _every_ page through WebCat ? (1997)
really wierd browser truncating (1997)
NT version and O'reily's WebSite (1997)
CASTEGORIES IN DIFFERENT FRAMES (1997)
E-mail Attachments (1997)
NT [delete] (1998)
Ship Cost Not working (2000)
WebCatalog can't find database (1997)
Locking up with WebCatalog... (1997)
3rd party processor question (2003)
Cancel Subscription (1996)
error: Too many nested [xxx] contexts (1997)
WebCat2b12 Mac.acgi--[searchstring] bug (1997)
Locking up with WebCatalog... (1997)
Mimeheaders parameters, which exist? (2003)
New Guestbook Source (1997)
WebCat2.0 [format thousands .0f] no go (1997)
return missing item (was:WebCat Sales) (1997)