client script to measure server performance: ideas ?

What I would like to do is to spawn multiple rebol processes that try to read a page from a local Apache server. If I can control the number of processes to spawn using "launch" and measure the response time of each read then I could measure the server's response time as a function of the number of simultaneous reads. Does anybody have any ideas or experience using launch ? What is the launch/link url for ? Thanks Jose _______________________________________________________________ Copa del Mundo de la FIFA 2002 El único lugar de Internet con vídeos de los 64 partidos. ¡Apúntante ya! en http://fifaworldcup.yahoo.com/fc/es/

I would recommend your script to launch, eg. 10 new processes, passing them a start time and a unique id number, one by one. How to pass the information across? Three ways: 1) save and read a file with the information in it 2) via a server/client exchange on a tcp port 3) write to and read from clipboard:// The launched rebol processes get this information so they know when to start. So they all wait for the start time then begin to make requests together. After finishing the attack (I mean, test) each process stores its results in a file with a filename corresponding to its id. (ie. results1, results2, ... results10) The original, mother script examines these files after an appropriate wait to collate the results. (Or I suppose you could use clipboard:// again to get the data serially). I am almost inspired to do this myself. Let me know how you go. Anton.

> What I would like to do is to spawn multiple rebol > processes that try to read a page from a local Apache > ? > What is the launch/link url for ?

I´m waiting to get more input so that I can understand the issues better . I imagine someone has thought about this for Rugby or another app Jose --- Anton <[anton--lexicon--net]> escribió: > I would recommend your script to launch, eg.

> 10 new processes, passing them a start time and > a unique id number, one by one. > [rebol-request--rebol--com] with "unsubscribe" in the > subject, without the quotes.

I tried it myself, but my results didn`t quite jive. They were much slower than expected. I believe it was because i was reading the pages. I think a better methodology would be to send a request for a page but not to process it. Instead of trying to spawn new processes inside of REBOL I did an shell script to repeatly call REBOL on the command line with a script. I`m willing to work with you to develop a better way. Anyone else got a better idea ? ----- Original Message ----- From: "jose" <[jjmmes--yahoo--es]> To: <[rebol-list--rebol--com]> Sent: Wednesday, June 05, 2002 10:58 AM Subject: [REBOL] client script to measure server performance: ideas ?

> What I would like to do is to spawn multiple rebol > processes that try to read a page from a local Apache > [rebol-request--rebol--com] with "unsubscribe" in the > subject, without the quotes.

This is not as easy as I thought ! See details on httperf C tool: For example, httperf is designed to run as a single-threaded process using non-blocking I/O to communicate with the server and with one process per client machine. http://www.hpl.hp.com/personal/David_Mosberger/httperf/doc004.html ========================================== 3 Design The two main design goals of httperf were (a) predictable and good performance and (b) ease of extensibility. Good performance is achieved by implementing the tool in C and paying attention to the performance critical execution paths. Predictability is improved by relying as little as possible on the underlying OS. For example, httperf is designed to run as a single-threaded process using non-blocking I/O to communicate with the server and with one process per client machine. With this approach, CPU scheduling is trivial for the OS which minimizes the risk of excessive context switching and poor scheduling decisions. Another example is timeout management: rather than depending on OS-mechanisms, httperf implements its own, specialized and light-weight timer management facility that avoids expensive system calls and POSIX signal delivery wherever possible. --- Adrian <[a--none--email--it]> escribió: > I tried it myself, but my results didn`t quite jive.

> They were much slower > than expected. I believe it was because i was > > > >

> > Copa del Mundo de la FIFA 2002 > > El único lugar de Internet con vídeos de los 64 > La qualità è garantita! > Clicca qui:

> -- > To unsubscribe from this list, please send an email > to > [rebol-request--rebol--com] with "unsubscribe" in the > subject, without the quotes. >

Hiya... Im not quite sure what your searching for? Server performance on HTML commands? of protocol speed? or even rebol read speed? (R)egards, Norman.

See my previous link mentioning httperf I'm looking for a script that can help me measure server performance by changing the client load and measuring the response times. The script has to be written in a way that can cleanly measure the server bottleneck without inteference from the client or the network. I´m looking for a httperf in rebol --- [rebolinth--nodep--dds--nl] escribió: >

> Hiya... > Im not quite sure what your searching for? Server > [rebol-request--rebol--com] with "unsubscribe" in the > subject, without the quotes.