Recently I observed a peculiar behaviour with Mellanox ConnectX-5 100 Gbps NIC. While working on 100 Gbps capture using DPDK Rxonly mode. It was observed that I was able to receive 142 Mpps using 12 queues. However with 11 queues, it was only 96 Mpps, with 10 queues 94 Mpps, 9 queues 92 Mpps. Can anyone explain why there is a sudden/abrupt jump in capture performance from 11 queues to 12 queues?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| How to find the maximum number of RX Queues for a NIC (ConnectX-5)? | 4 | 5279 | April 7, 2021 | |
| ConnectX-5 RX RSS has no impact even allocating more rx queues | 2 | 1128 | March 29, 2021 | |
| ConnectX-4 RX performance issues on DPDK | 2 | 1475 | March 25, 2021 | |
| all my mlx cx5 2*100G cards unable to line rate speed | 1 | 1277 | September 9, 2021 | |
| Configuring NIC with 0 RX queue (TX queues only) on DPDK 21.11 | 4 | 2401 | March 7, 2022 | |
| Mellanox ConnectX-5 Tx performance issue | 1 | 1236 | May 19, 2022 | |
| Performance Test finding bottleneck and optimization | 2 | 1983 | March 17, 2022 | |
| I use the AF_XDP and the example of the linux kernel (xdpsock) and I reach a maximum at 1Mpps in Tx and 5Mpps in Rx (and zero-copy is not activable when it should be I think), where I would be expecting 15/20Mpp | 1 | 473 | January 21, 2020 | |
| Significant performance issue with MPRQ on ConnectX-6 / DPDK 23.11 | 3 | 1029 | February 19, 2024 | |
| DPDK rte_flow is degrading performance when testing on Connect X5 100G EN @ 100G | 7 | 1929 | September 25, 2025 |