bytedance/byteps

How did you get the horovod & bytePS performance

Open

#68 opened on Jul 18, 2019

View on GitHub
 (12 comments) (2 reactions) (0 assignees)Python (3,483 stars) (476 forks)batch import
distributeddocumentationgood first issue

Description

I have the same hardware envs, same network, but I could not get the result as you, almost half as you. Any best practices and experience? thanks very much! for bytePS with 1 instance and 8 GPU, I have similar testing result.

Contributor guide

Tech stack
python
Domain
machine learningperformance
Issue type
performance
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
over 1 week
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
unclear
Prerequisites
distributed trainingBytePSHorovodGPU cluster
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
10
Research direction
The issue is a open ended query about reproducing BytePS performance benchmarks. To assist, first clarify the user's exact hardware, software versions, and network topology. Then, review BytePS configuration files and compare with published benchmarks. Check for common pitfalls like NCCL settings, network latency, or GPU affinity. Provide step by step tuning guidance and recommend running a simple allreduce test to isolate the bottleneck.