Marc,

You should be aware that SIPp does have some hidden limitations regarding
the length of a run.  There are 32-bit quantities that are used to express
time, so that means that there are only about 4 billion addressable
milliseconds.  This works out to 49 days and 17 hours before things wrap
around.

Olivier, you should be aware that I made use of this fact in my runqueue
patch and divided time up into three timing wheels (10 bits, 10 bits, and
12 bits if I recall) that address only 32 bits.  Addressing more bits would
not be a problem, but a fourth wheel should be added and the bits would
need to be redistributed among the wheels.  The same issue would also
affect any move to sub-millisecond timer resolution.

Charles



                                                                           
             <[EMAIL PROTECTED]                                             
             elgacom.be>                                                   
             Sent by:                                                   To 
             sipp-users-bounce         <[EMAIL PROTECTED]>               
             [EMAIL PROTECTED]                                          cc 
             ge.net                    [email protected]    
                                                                   Subject 
                                       [Sipp-users] defunct solution OK +  
             11/14/2006 08:22          small correction patch for          
             AM                        flush(screenf)                      
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           




Hi Olivier,


finally found some time to look at sipp again, and have the version
sipp.2006-11-08 running some hours now with quite a lot of "exec"'s in it.
All seems OK, no defuncts any more and I don't notice any severe
performance issues.


I did notice some time ago that the screen log file where you can get the
nice info from interactive mode when running in background, did not
properly write the results to the file, i.e. it often missed the last part.
So I added the flush to the code version sipp.2006-11-08, patch file in
attach.


<<flush_diff.txt>>
For information and maybe publicity: I've got a sipp instance running now
for over 20 days, with more than a million messages sent over some 400
ports.





Best regards,


  MarcVD


(-: from Marc VAN DIEST (BELGACOM ANS/NTA) +32 2 244 5078 ;-)


                                                                            
 **** DISCLAIMER **** http://www.belgacom.be/maildisclaimer                 
                                                                            


(See attached file: flush_diff.txt)
-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job
easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Sipp-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sipp-users

Attachment: flush_diff.txt
Description: Binary data

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Sipp-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sipp-users

Reply via email to