bytedance/deer-flow

发现每个沙箱要占用将近1个g内存啊，这咋做高并发，有没有好主意~

Open

#3,213 opened on May 25, 2026

View on GitHub
 (7 comments) (0 reactions) (0 assignees)Python (67,767 stars) (9,005 forks)batch import
help wanted

Description

NAME CPU(cores) MEMORY(bytes)
deerflow-backend-674c55c457-dqzjx 14m 395Mi
postgres-0 6m 68Mi
provisioner-776b74bc67-47grq 3m 91Mi
sandbox-37060a71 29m 792Mi
sandbox-385fe401 36m 501Mi
sandbox-c39325de 32m 582Mi
sandbox-f0036078 25m 828Mi

Contributor guide

Tech stack
pythondocker
Domain
performance
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.
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
PythonDockerMemory profilingSandbox concept
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
35
Research direction
Investigate the memory footprint of sandbox containers. Start by profiling the sandbox Python process using tools like memory profiler to identify memory heavy objects or leaks. Check the Dockerfile and entrypoint scripts in the sandbox directory (e.g., sandbox/Dockerfile) to see if there are unnecessary dependencies or large base images. Consider using shared memory or lightweight sandboxing techniques. Review comments for any suggestions from other contributors.