alibaba / alibaba/x-deeplearning

如何修改才能支持PS,Scheduler,Worker都在特定的ip+port启动?

Open
#315 0 comments 1 reaction 0 assignees View on GitHub
Dominant language
PureBasic
Stars
4.3k
Forks
1k
PR merge metrics
No merged PRs in 30d

Description

1. 了解的背景:
xdl 目前启动server和worker都是自动寻找可用的端口+获取默认IP,启动seastar作为底层通讯工具。
2. 需求:
公司内集群调度docker分配给每台ps,scheduler,worker特定的ip和端口号,意思是只能在特定的ip+port下才能建立通讯连接,请问需要修改哪些地方才可以满足此项需求呢?
3. 进度:
(1) 修改完毕ps-plus中scheduler和server相关ip和port启动参数
(2)client 端(worker) 如何 只通过特定端口与scheduler通信呢?(建立的EventClientLib好像实现相应功能?)
真心求教,谢谢。

Contributor guide

No contributing guide indexed for this repository

Research direction

Review the existing ps-plus scheduler and server IP/port startup changes, then inspect EventClientLib and the worker-to-scheduler connection path. Identify the configuration or startup entry points controlling binding and client destinations, and verify that PS, scheduler, and workers can communicate using their assigned addresses.

Written by the indexing model from the issue text.

Assessment

Tech stack
docker
Domain
distributed-systems, networking
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.