This week's book giveaway is in the Cloud/Virtualization forum.
We're giving away four copies of Pipeline as Code and have Mohamed Labouardy on-line!
See this thread for details.
Win a copy of Pipeline as Code this week in the Cloud/Virtualization forum!

neil upfalow

Greenhorn
+ Follow
since Jun 02, 2010
Cows and Likes
Cows
Total received
0
In last 30 days
0
Total given
0
Likes
Total received
0
Received in last 30 days
0
Total given
0
Given in last 30 days
0
Forums and Threads
Scavenger Hunt
expand Ranch Hand Scavenger Hunt
expand Greenhorn Scavenger Hunt

Recent posts by neil upfalow

i'm curious about this too for a slightly different reason. one thing i noticed is you have timeout of 20000.
most tuning topics n vids indicate if apache is front ending the requests that you can bring the timeout down to say 300...since there is no reason to maintain the connection as apache is doing that w the browser. perhaps you won't see threads dying until that 20000secs have elapsed....whic his a pretty long time. 300s is 5m.

i do see our tomcat 5.5 systems reduce threadcount. To me it appears in a monitorng program that shows tomcat threads as processes.

if anyone knows, i'd love to know an answer i have on another topic i started about threads/procs. I'd like to run no more than are needed. If the hardware only has 24 cores and the normal request work is done n complete within 100ms or so, what do i need more than even 24 threads / procs? I ask because with many VEs running tomcat the total threads grows to a ridiculous number and very few seem to accumulate cpu time with the rest just always in a sleeping state.
10 years ago
anyone know what the benefits are on of having more threads than cpu cores given they execute and finish their work very fast?
10 years ago
Is there any reason to have higher MaxThreads that the number of cpu cores you have?
For example, say there are virtual servers running on a 4 cpu (6cores ea) system. They have traffic that's low, usually a few req/s and mostly for httpd apache anyway. Tomcat jsps n beans execute really fast.
If no virtual server's tomcat could run on more than 24 cores simultaneously, why would i want more than 24 threads as Max.
In reality, it seems apache has 1-2s of work to do after a jsp takes about 100ms to do its total work. I never see more than 20-30 httpd processes running and jsps are so much faster, seems tomcat would not even need 24 threads to back end apaches requests to it.

If i had a tomcat thread max for 100 and a min if 50 ....wouldn't all but 24 threads never even have a chance of executing?
Are threads used for anything else and if so how many?

I would like to pursue a major reduction of total processes running on the hardware and most are tomcat and almost all of them are, in reality, sleeping most of the time....so it seems to me i'd never need more than 24 max.

I'm also curious about reducing threads that need to be killed so if 24 max makes sense for 24 cores then i'd just set min n max to 24 so no cpu time is wasted on thread creation nor destruction.

thanks in advance!
10 years ago
I agree. I think maybe the translet ability of xsltc mimics the ability of XT to create a sort of compiled representation of the xsl rules for faster exec and lower ram. I don't know that it is any faster though and maybe that is just what xt does behind the scenes.

I'm still hoping to know if anyone has done both and knew a rough comparison of performance before i embark on the ramp up, dev and perf testing etc.

if i do it in the next week or so i will post the results.
Hi,
Is newer xstl engine faster than jclark XT ?
I would like to hear from anyone with excellent knowledge of the current xslt benchmarks and components for java.
We are currently using XT. Based on a setting, it evaluates the xsl fresh ever execution and creates a compiled version (like when we are editing/refining xsls) or the setting stops evaluation and uses the last generated compiled xsl files ( for production mode performance) . this is a webstore scenario in tomcat environment distributed to a many installations on various clients virtual dedicated servers where performance limits are controlled and limited. XSL transforms account for most of the cpu workload. I'd like to save cpu time for the users by improving what is the slowest mechanism.

The XSLs are relatively small and simple xsls (written with performance in mind) to produce parts of each page, 5 or so to make up an entire page. Some only take a few millis to run. some have output that is even cached as it changes infrequently. however, some like category type pages that must loop perhaps 200 items with some minimal if/when logic here and there could take 200ms. We'd love to cut that in half or more.

Can newer xslt engines run significantly faster, given this type of scenario, and if so which?

All the benchmarks i have looked at are really outdated from back in 2000 timeframe and with early versions of xsltc. My own tests back in 2003-2004 confirmed XT was way faster than anything i tested back then.

thanks in advance.