FedML-AI/FedML

GRPC communication to slow

Open

#234 opened on May 29, 2022

View on GitHub
 (1 comment) (0 reactions) (3 assignees)Python (743 forks)batch import
TODO featureshelp wanted

Repository metrics

Stars
 (3,930 stars)
PR merge metrics
 (No merged PRs in 30d)

Description

I'm doing SplitNN experiment today on LAN through JET TX2 development board and Ubunut 20.04 server, transferring about 100M featuremap, the transfer time is 17 seconds ...... our LAN is 1000M transfer ...... want to ask why so slow

Contributor guide

Research direction
Profile the GRPC communication and compare actual throughput against theoretical LAN speed. Check message size, serialization overhead, and GRPC configuration (e.g., flow control, compression).
Tech stack
pythongrpc
Domain
backendperformance
Issue type
Performance
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonGRPCLinux
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
45