From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 6EB3E4659A for ; Tue, 15 Apr 2025 16:22:25 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 6443640289; Tue, 15 Apr 2025 16:22:25 +0200 (CEST) Received: from szxga04-in.huawei.com (szxga04-in.huawei.com [45.249.212.190]) by mails.dpdk.org (Postfix) with ESMTP id 2476840263 for ; Tue, 15 Apr 2025 16:22:24 +0200 (CEST) Received: from mail.maildlp.com (unknown [172.19.162.112]) by szxga04-in.huawei.com (SkyGuard) with ESMTP id 4ZcRF12N3Cz2TS5W; Tue, 15 Apr 2025 22:22:13 +0800 (CST) Received: from dggpemf200011.china.huawei.com (unknown [7.185.36.152]) by mail.maildlp.com (Postfix) with ESMTPS id DD70B140143; Tue, 15 Apr 2025 22:22:19 +0800 (CST) Received: from kwepemo100017.china.huawei.com (7.202.195.215) by dggpemf200011.china.huawei.com (7.185.36.152) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.11; Tue, 15 Apr 2025 22:22:19 +0800 Received: from kwepemo100017.china.huawei.com ([7.202.195.215]) by kwepemo100017.china.huawei.com ([7.202.195.215]) with mapi id 15.02.1544.011; Tue, 15 Apr 2025 22:22:19 +0800 From: "jiangheng (G)" To: Matan Azrad , Slava Ovsiienko , "users@dpdk.org" Subject: mellanox CX5 100GbE NIC performance test on X86 and ARM server Thread-Topic: mellanox CX5 100GbE NIC performance test on X86 and ARM server Thread-Index: AduuEZJfFdONyKOgR4eVnZO7j18+tA== Date: Tue, 15 Apr 2025 14:22:19 +0000 Message-ID: Accept-Language: en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.136.117.195] Content-Type: multipart/alternative; boundary="_000_a9c2f7e9e8f64e0bb74deece43607615huaweicom_" MIME-Version: 1.0 X-BeenThere: users@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK usage discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: users-bounces@dpdk.org --_000_a9c2f7e9e8f64e0bb74deece43607615huaweicom_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi, I am testing the perfomance of the mellanox CX5 100GbE NIC. I found that th= e perfomance(Mpps) of AMD(X86) server is better than that of Kunpeng920 Ser= ver(ARM) in the single-core scenario. the test command is as follows: RX-side: dpdk-testpmd -l 1-23 -n 4 -a XXXX -- -i --rxq=3D1 --txq=3D1 --txd=3D1024 --= rxd=3D1024 --nb-cores=3D1 --eth-peer=3D0,xxxx --burst=3D128 --forward-mode= =3Drxonly -a --txpkts=3D128 --mbcache=3D512 --rss-udp TX-side (payload_size is 128byte): dpdk-testpmd -a XXXXX -l 1-23 -n 4 -- -i --rxq=3D4 --txq=3D4 --txd=3D1024 -= -rxd=3D1024 --nb-cores=3D4 --eth-peer=3D0,XXXXX --burst=3D64 --forward-= mode=3Dtxonly -a --txpkts=3D128 --mbcache=3D512 --rss-udp firmware-version: 16.32.1010 (HUA0000000004) OS: OpenEuler 22.03 Kernel: 5.10 DPDK: 21.11.5 Results: ARM: 28.598Gbps, 27.928Mpps X86: 34.015Gbps, 33.218Mpps After some checks, I suspect that the bottleneck is mainly the NIC. Have yo= u tested the performance of the CX5 on the ARM server? Do you have any optimization methods for ARM server, such as some parameter= s or firmware versions? Thanks --_000_a9c2f7e9e8f64e0bb74deece43607615huaweicom_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hi,

I am testing the perfomance of = the mellanox CX5 100GbE NIC. I found that the perfomance(Mpps) of AMD(X86) = server is better than that of Kunpeng920 Server(ARM) in the single-core sce= nario. the test command is as follows:

RX-side:

dpdk-testpmd -l 1-23 -n 4 -a XX= XX -- -i --rxq=3D1 --txq=3D1 --txd=3D1024 --rxd=3D1024 --nb-cores=3D1 --eth= -peer=3D0,xxxx --burst=3D128 --forward-mode=3Drxonly -a --txpkts=3D12= 8 --mbcache=3D512 --rss-udp

TX-side (payload_size is 128byt= e):

dpdk-testpmd -a XXXXX -l 1-23 -= n 4 -- -i --rxq=3D4 --txq=3D4 --txd=3D1024 --rxd=3D1024 --nb-cores=3D4 = ; --eth-peer=3D0,XXXXX --burst=3D64 --forward-mode=3Dtxon= ly -a --txpkts=3D128 --mbcache=3D512 --rss-udp

firmware-version:

16.32.1010 (HUA0000000004)=

OS:

OpenEuler 22.03

Kernel:

5.10

DPDK:

21.11.5

Results:

ARM:

28.598Gbps, 27.928Mpps

X86:

34.015Gbps, 33.218Mpps

After some checks, I suspect th= at the bottleneck is mainly the NIC. Have you tested the performance of the= CX5 on the ARM server?

Do you have any optimization me= thods for ARM server, such as some parameters or firmware versions?

Thanks

--_000_a9c2f7e9e8f64e0bb74deece43607615huaweicom_--