[an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] (none) [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive] (none) [an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive][an error occurred while processing this directive]
 
[an error occurred while processing this directive] [an error occurred while processing this directive]
Skåne Sjælland Linux User Group - http://www.sslug.dk Home   Subscribe   Mail Archive   Forum   Calendar   Search
MhonArc Date: [Date Prev] [Date Index] [Date Next]   Thread: [Date Prev] [Thread Index] [Date Next]   MhonArc
 

Re: [PROGRAMMERING] Hvordan ? Et program skal sende beskeder til N andre.



> Intressant att msg2 ofta är lite snabbare än msg1, kan det förklaras med
> att många processer väntar på samma kö?
En grund kunne være, at msq1 modtagerne skal filtrere i beskederne,
hvis scheduleringrækkefølgen ikke er child 1,2,3 ... n, parent, 1,2,3 ... n, parent, 1,2 ..
En anden kunne være, at kernen er længere om at håndtere en længere kø.

Der er andre bemerkelsesværdige ting ved testen:
1 - pthread (user + kernel) er bedre end msqX, men pthread (wall) er dårligere. Linux straffer single process modellen?
2 - msqX løsningerne brænder en forholdsvis stor andel af løbetiden af i kernen. Både pthread og shm1 bruger en del mindre
    kernel time
3 - shm1 er i gennemsnit hurtigst med en kø på 2 beskeder. Man skulle tro at det var bedst med en større kø, da
    man undgår nogle context switches på den måde. Her kunne man nok vælge en mere optimal schedulerindstilling.

Og så synes jeg jo, at shm2 burde virke. Det er mystisk, hvorfor den oftest går ned.

> med typen msg2, eftersom jag redan har en enkel infrastruktur kring det,
> och ändå vara snabb.
Ja, også fordi testen stort set kun måler kopieringstiden. Med en ordentlig stor beskedshåndteringstid bliver forholdet
bedre for msqX. Men for msq2 og mange (broadcast) modtagere ser det ud til, at man skal øge max antallet af køer, som
kernen understøtter.

Hans-Christian Stadler



 
Home   Subscribe   Mail Archive   Index   Calendar   Search

 
 
Questions about the web-pages to <www_admin>. Last modified 2005-08-10, 22:44 CEST [an error occurred while processing this directive]
This page is maintained by [an error occurred while processing this directive]MHonArc [an error occurred while processing this directive] # [an error occurred while processing this directive] *