pytorch/serve

Enabling HTTP caching of inference results

Open

#1,527 opened on Mar 23, 2022

View on GitHub
 (2 comments) (4 reactions) (1 assignee)Java (790 forks)batch import
enhancementhelp wanted

Repository metrics

Stars
 (3,844 stars)
PR merge metrics
 (No merged PRs in 30d)

Description

Is your feature request related to a problem? Please describe.

Currently, TorchServe adds headers that prevent from caching the inference results: https://github.com/pytorch/serve/blob/30f83500b0850e26ec55581f48a9307b1986f9f9/frontend/server/src/main/java/org/pytorch/serve/util/NettyUtils.java#L187-L190

This prevents some reverse-proxies like nginx from caching the results

Describe the solution

It would be great to have a way to override this behavior, either from the model handler, or through some configuration key.

Describe alternatives solution

N/A

Contributor guide

Research direction
Investigate the file frontend/server/src/main/java/org/pytorch/serve/util/NettyUtils.java, specifically lines 187-190 where caching prevention headers are set. Consider adding a configuration option or model handler method to allow overriding these headers. Review existing configuration mechanisms in TorchServe to understand how to implement a toggle for caching. Look at discussions in the issue comments for any additional context or proposed solutions.
Tech stack
javapython
Domain
backendperformance
Issue type
Feature
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.
Half day
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.
Clear
Prerequisites
JavaNettyTorchServe internalsHTTP caching concepts
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
40