On Saturday 02 December 2006 14:26, Csibra Gergo wrote: > Well, the data bandwidth is only one. The irq is the other, and that > is the bottleneck.
Unless I am mistaken (first time for everything, <g>), the mutli-span cards issue 1 interrupt per every millisecond, and *all* spans on that card are serviced in that interrupt. To put it another way: TDM4xxP - 1-4 channels, 1000Hz interrupt rate TE100P - 1-24 channels, 1000Hz interrupt rate TE2xxP - 1-48 channels, 1000Hz interrupt rate TE4xxP - 1-96 channels, 1000Hz interrupt rate I know that multiple cards do not share the zaptel interrupt, so with two cards you have two 1000Hz interrupt sources. I have a back-burner "I wonder if" kind of idea which involves zaptel shutting down the interrupt on the 2nd (and more) cards and servicing all cards in one interrupt, but the back burner's so far away now I don't know if I'll ever get to it. :-) -A. _______________________________________________ --Bandwidth and Colocation provided by Easynews.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
