Hello, I am not sure to understand the mechanism behind dpdk concerning the context switches. I have two user space applications that need access to the NIC according to incoming port rules (port 80 and port 443).
How to be sure that DPDK spreads the load to the right application ? Will 2 dpdk instances be needed (one per app) -> two incoming packets analysis to "know" if the packet should be forwarded to the user space process ? Which would basically be the same thing as inefficient promiscuous mode. i don't understand that "filtering" point. Could you please clarify ? Thanks --? Jog