Speedup Multi-Core¶
Speedup Multi-Core throughput graphs are generated by multiple executions of the same performance tests across physical testbeds hosted LF FD.io labs: 2n-skx, 3n-skx, 2n-clx, 3n-tsh, 2n-tx2, 2n-dnv, 3n-dnv. Grouped bars illustrate the 64B/78B packet throughput speedup ratio for 2- and 4-core multi-threaded VPP configurations relative to 1-core configurations.
Additional information about graph data:
Graph Title: describes tested packet path, testbed topology, processor model, NIC model, packet size used by data plane workers and indication of VPP DUT configuration.
X-axis Labels: number of cores.
Y-axis Labels: measured Packets Per Second [pps] throughput values.
Graph Legend: lists CSIT test suites executed to generate graphed test results.
Hover Information: lists number of runs executed, specific test substring, mean value of the measured packet throughput, calculated perfect throughput value, difference between measured and perfect values and relative speedup value.
Note
Test results are stored in build logs from FD.io vpp performance job 2n-skx, build logs from FD.io vpp performance job 3n-skx, build logs from FD.io vpp performance job 2n-clx, build logs from FD.io vpp performance job 2n-zn2, build logs from FD.io vpp performance job 3n-tsh, build logs from FD.io vpp performance job 2n-tx2, build logs from FD.io vpp performance job 2n-dnv and build logs from FD.io vpp performance job 3n-dnv with RF result files csit-vpp-perf-2106-*.zip archived here. Required per test case data set size is 10, but for VPP tests the actual size varies per test case and is <=10.
- L2 Ethernet Switching
- IPv4 Routing
- IPv6 Routing
- SRv6 Routing
- IPv4 Tunnels
- NAT44 IPv4 Routing
- KVM VMs vhost-user
- LXC/DRC Container Memif
- IPSec IPv4 Routing