ConnectX-5 RX RSS has no impact even allocating more rx queues

I am running pktgen in another server, connected to RX server via 100G DAC (only one 100G port is used for testing). pktgen is generating 25 Mpps.

But at RX server, it is receiving at a rate of 12-14 Mpps. I tried RSS and spread it to 4 RX Queues and dedicated lcore for reading from each RX Queue. But the collective RX capability still remains 12-14 Mpps. No matter how many more Queues i inrease, total RX rate remains the same as 12-14 Mpps.

Any help would be highly appreciated. RSS conf I used at receiver side is given below

.rx_adv_conf = {

.rss_conf = {

.rss_hf = ETH_RSS_IP | ETH_RSS_UDP |




Here is a sample output where I am priting no:of pkts received from each RX queue read within a second. Total 4 Queues

Instance: 0 . Pkts: 3104552

Instance: 1 . Pkts: 3015873

Instance: 2 . Pkts: 3798675

Instance: 3 . Pkts: 2729060

IN_PKTS Total: 12648174

Hello Prem,

Thank you for posting your inquiry on the NVIDIA Networking Community.

Based on the information provided, the issue is not related to a malfunction of the adapter or the mlx5 PMD driver.

Please review the latest DPDK performance reports related to tuning and testing DPDK to achieving full performance on our adapters.

Thank you and regards,

~NVIDIA Networking Technical Support

Hello Martijn,

Thank you for the reply. I will go through the performance optimizations for dpdk with mellanox card.

Since i have not tried any cards other than connect-x-5 , just need confirmation for below two points.

  1. Can Mellanox card will be able to intake more packets (pps) if we increase the RX queues using RSS capability.
  2. I am using AMD Epyc 7702 processor (64 cores). Does using AMD proc can impact the performance with mellanox card? Should I consider intel xeon processor ?

Thanks & Regards,

Prem Sankar