> > One example is network traffic. This form of activity > > generates large amounts of device interrupts (from the > > network interfaces) which puts a *huge* burden on both > > the system and the scheduler. Adaptation to massive > > network load would simply be to "place the core thread > > which handles network traffic onto the main CPU so > > that it lessens network latency and can handle the > > device interrupt and process the packet without having > > to wait until another TCP/IP thread is awoken on > > another processor to handle the request". The daemon > > I believe the new "FireEngine" TCP/IP stack in Solaris 10 > already does this. I know that certianly some processor > affinity goes on.
yes, work was done in this area as part of fire engine and as part of nemo (our new network driver framework), and we currently do dynamic interrupt blanking -- check out the SQS_POLL_CAPAB handling in squeue_drain(). we have *a lot* more upcoming work in this area underway with the crossbow project, which sunay can fill interested folks in on (on the networking-discuss alias). -- meem _______________________________________________ opensolaris-code mailing list [email protected] https://opensolaris.org:444/mailman/listinfo/opensolaris-code
