Soft Patch Panel
 help / color / mirror / Atom feed
From: x-fn-spp-ml@ntt-tx.co.jp
To: spp@dpdk.org, yasufum.o@gmail.com
Subject: [spp] [PATCH v2 5/5] docs: add note regarding NUMA configuration
Date: Wed, 24 Jun 2020 13:47:36 +0900	[thread overview]
Message-ID: <20200624044736.14540-6-x-fn-spp-ml@ntt-tx.co.jp> (raw)
In-Reply-To: <20200529110932.26508-1-x-fn-spp-ml@ntt-tx.co.jp>

From: Hideyuki Yamashita <yamashita.hideyuki@ntt-tx.co.jp>

This patch adds new notes for NUMA configuration when starting
spp_primary.

This is pointed out by the following DPDK Bugzilla tickets.
https://bugs.dpdk.org/show_bug.cgi?id=457

Signed-off-by: Hideyuki Yamashita <yamashita.hideyuki@ntt-tx.co.jp>
Signed-off-by: Naoki Takada <ntakada14@gmail.com>
---
 docs/guides/gsg/performance_opt.rst | 55 +++++++++++++++++++++++++++++
 1 file changed, 55 insertions(+)

diff --git a/docs/guides/gsg/performance_opt.rst b/docs/guides/gsg/performance_opt.rst
index 6fea4cb..24436a5 100644
--- a/docs/guides/gsg/performance_opt.rst
+++ b/docs/guides/gsg/performance_opt.rst
@@ -71,6 +71,61 @@ and so on.
     pid 192626's current affinity list: 0-31
     pid 192626's new affinity list: 7
 
+Consideration of NUMA node
+--------------------------
+
+``spp_primary`` tries to create memory pool in the same NUMA node where
+it is launched. Under NUMA configuration, the NUMA node where ``spp_primary``
+is launched and the NUMA node where NIC is connected can be different
+(e.g. spp_primary runs in NUMA node 0 while NIC is connected with NUMA
+node 1).
+Such configuration may cause performance degradation. In general, under
+NUMA configuration, it is best practice to use CPU and NIC which belongs
+to the same NUMA node for best performance. So user should align those
+when performance degradation makes the situation critical.
+
+To check NUMA node which CPU/NIC core belongs, ``lstopo`` command can be used.
+In the following example, CPU core 0 belongs to NUMA node 0 while
+``enp175s0f0`` belongs to NUMA node 1.
+
+.. code-block:: console
+
+    $ lstopo
+    Machine (93GB total)
+      NUMANode L#0 (P#0 46GB)
+        Package L#0 + L3 L#0 (17MB)
+          L2 L#0 (1024KB) + L1d L#0 (32KB) + L1i L#0 (32KB) + Core L#0
+    .....
+      NUMANode L#1 (P#1 47GB)
+        Package L#1 + L3 L#1 (17MB)
+          L2 L#12 (1024KB) + L1d L#12 (32KB) + L1i L#12 (32KB) + Core L#12
+            PU L#24 (P#1)
+            PU L#25 (P#25)
+    .....
+        HostBridge L#10
+          PCIBridge
+            PCI 8086:1563
+              Net L#10 "enp175s0f0"
+            PCI 8086:1563
+              Net L#11 "enp175s0f1"
+
+CPU core where ``spp_primary`` run can be specified
+using -l option.
+
+.. code-block:: console
+
+    # terminal 3
+    $ sudo ./src/primary/x86_64-native-linux-gcc/spp_primary \
+        -l 0 -n 4 \
+        --socket-mem 512,512 \
+        --huge-dir /dev/hugepages \
+        --proc-type primary \
+        --file-prefix $SPP_FILE_PREFIX \
+        --base-virtaddr 0x100000000
+        -- \
+        -p 0x03 \
+        -n 10 \
+        -s 192.168.1.100:5555
 
 Reference
 ---------
-- 
2.17.1


      parent reply	other threads:[~2020-06-24  4:47 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20200529110932.26508-1-x-fn-spp-ml@ntt-tx.co.jp>
2020-06-24  4:47 ` [spp] [PATCH v2 0/5] updates for catching up DPDK20.02 x-fn-spp-ml
2020-06-30  6:48   ` Itsuro ODA
2020-07-16 15:56   ` Yasufumi Ogawa
2020-06-24  4:47 ` [spp] [PATCH v2 1/5] docs: remove unnecessary usage of ethtool x-fn-spp-ml
2020-06-24  4:47 ` [spp] [PATCH v2 2/5] docs: resolve broken reference link x-fn-spp-ml
2020-06-24  4:47 ` [spp] [PATCH v2 3/5] docs: catch up updates in dpdk-20.02 x-fn-spp-ml
2020-06-24  4:47 ` [spp] [PATCH v2 4/5] makefile: change default build target x-fn-spp-ml
2020-06-24  4:47 ` x-fn-spp-ml [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200624044736.14540-6-x-fn-spp-ml@ntt-tx.co.jp \
    --to=x-fn-spp-ml@ntt-tx.co.jp \
    --cc=spp@dpdk.org \
    --cc=yasufum.o@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).