Help us understand the problem. What is going on with this article?

AMD Polaris(RX570)でのRCCL動作検証

More than 1 year has passed since last update.

なんといつのまにかRCCLがapt installできるようになってました

環境構築

sudo apt install rccl

ROCmがインストールされている環境なら割とすぐに立ち上げることができます。

$ git clone https://github.com/ROCmSoftwarePlatform/rccl-tests.git
$ cd ./rccl-tests/
$ sudo ./install.sh 

これでrccl-testのビルドが完成し

./rccl-test/build内にビルドされます

2wayの場合

./build/all_reduce_perf -b 8 -e 128M -f 2 -g 2

でテストできます。

動作検証結果

 ./build/all_reduce_perf -b 8 -e 128M -f 2 -g 2

# nThread: 1 nGpus: 2 minBytes: 8 maxBytes: 134217728 step: 2(factor) warmupIters: 5 iters: 20 validation: 1 
#
# Using devices
#   Rank  0 Pid   9012 on       rocm device  0 [0x02] Ellesmere [Radeon RX 470/480/570/570X/580/580X]
#   Rank  1 Pid   9012 on       rocm device  1 [0x03] Ellesmere [Radeon RX 470/480/570/570X/580/580X]
#
#                                                     out-of-place                       in-place          
#       size         count    type   redop     time   algbw   busbw  error     time   algbw   busbw  error
#        (B)    (elements)                     (us)  (GB/s)  (GB/s)            (us)  (GB/s)  (GB/s)       
Memory access fault by GPU node-2 (Agent handle: 0x1d58110) on address (nil). Reason: Page not present or supervisor privilege.
中止 (コアダンプ)

残念ながらテストは完走できませんでした
メモリアクセス障害となっているので改善が必要なやつかもしれません。

環境

OS ubuntu 16.04.6 64bit
kernel 4.15.0-58-generic
ROCm version R2.7
RCCL version 2.6.0.182-rocm-rel-2.7-22-1fee6f9

CPU Intel® Xeon(R) CPU E5-2603 v4 @ 1.70GHz × 6
MB X99S GAMING 7
RAM DDR4 2400 16GBモジュール*3 48GB
GPU0 RX570 16GB
GPU1 RX570 16GB

T_keigo_wwk
記事の多くはメモ、当方での動作検証結果がメインで現時点での再現性を保証するものではありません。 引用する際は引用先を明示して行うよう心がけておりますが抜けている箇所、事実と異なるや技術的問題等ある箇所がありましたら御一報ください
Why not register and get more from Qiita?
  1. We will deliver articles that match you
    By following users and tags, you can catch up information on technical fields that you are interested in as a whole
  2. you can read useful information later efficiently
    By "stocking" the articles you like, you can search right away
Comments
No comments
Sign up for free and join this conversation.
If you already have a Qiita account
Why do not you register as a user and use Qiita more conveniently?
You need to log in to use this function. Qiita can be used more conveniently after logging in.
You seem to be reading articles frequently this month. Qiita can be used more conveniently after logging in.
  1. We will deliver articles that match you
    By following users and tags, you can catch up information on technical fields that you are interested in as a whole
  2. you can read useful information later efficiently
    By "stocking" the articles you like, you can search right away
ユーザーは見つかりませんでした