At 06:35 PM 2/27/00 -0500, jamal wrote:
>
>
>On Mon, 28 Feb 2000, Dennis wrote:
>
>> At 06:42 PM 2/26/00 -0500, jamal wrote:
>> >
>> >
>> >>We have a customer who has peaked at 110k pps on dual T3s on a 600Mhz
>> >>Pentium. It happens to be FreeBSD but I suspect that a linux number would
>> >
>> >Out of curiosity, how was the 110Kpps measured?
>>
>> Interrupts generated by the HSSI card.
>>
>
>
>So a more stupid precise question:
>How did you count it to get the interupts/sec?
>Ie did you take a long time average or did you use some little tool
>which gave you a more accurate peaks. I can understand if you tested with
>some piece of hardware like the smartbits, but from your context it seems
>to be from real life traffic
> Seems like 110Kpps is already beyond T3 capacity.
*yawn* Ever hear of vmstat -i (or /proc/interrupts in linux-speak?)
Lets see, 90Mb/s / (40 * 8) = over 281,000pps max per HSSI, and I said dual
HSSIs so thats 560,000pps possible.
at least try to do the math before wasting bandwidth on the list please.
Dennis
-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]
