Re: [c-nsp] NPE-G2: mixed interrupt/polling packet processing on POS interface

2008-02-25 Thread Rodney Dunn
For G2 vs G1 that's correct. For ethernet vs. serial the performance from what I recall is less. Rodney On Thu, Feb 21, 2008 at 12:35:32PM +0100, Ultramajestic wrote: So as far as I understand, the performance is not degraded, isn't it? El mi??, 20-02-2008 a las 09:30 -0500, Rodney Dunn

Re: [c-nsp] NPE-G2: mixed interrupt/polling packet processing on POS interface

2008-02-21 Thread Ultramajestic
So as far as I understand, the performance is not degraded, isn't it? El mié, 20-02-2008 a las 09:30 -0500, Rodney Dunn escribió: Philippe, I need to check back with the BU on it. I thought they were putting a document out on this G2 processing reporting issue. Basically due to the way

Re: [c-nsp] NPE-G2: mixed interrupt/polling packet processing on POS interface

2008-02-20 Thread Rodney Dunn
Philippe, I need to check back with the BU on it. I thought they were putting a document out on this G2 processing reporting issue. Basically due to the way the G2 does CPU accounting it looks like it's higher at lower loads based on the CPU measurements. If you put it in the lab and measure