Doc fix for Issue 860, warn about OOME when building Docker image

[fbbradheintz]

Description

When creating a Docker image that builds TorchServe from source, one needs at least 4GB of RAM allocated to the container. If too little RAM is allocated, OOMEs cause installation (specifically pip's installation of PyTorch, the package for which is >700MB) to fail with an uninformative error message ("Killed").

This PR adds a note to the documentation for setting up the KFServing Docker image recommending a 4GB minimum RAM allocation.

Fixes #860

Type of change

This change is a documentation update

Feature/Issue validation/testing

The instructions to build the Docker image fail (on my Mac running macOS 10.15.7) if too little RAM is allocated, but succeed consistently when there's sufficient memory. This has been tested multiple times.

Checklist:

• n/a Have you added tests that prove your fix is effective or that this feature works?
• n/a New and existing unit tests pass locally with these changes?
• n/a Has code been commented, particularly in hard-to-understand areas?
• [√] Have you made corresponding changes to the documentation?

[fbbradheintz]

cc @maaquib @chauhang @jeremiahschung

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-cpu
• Commit ID: 75940f4
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-win
• Commit ID: 75940f4
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-gpu
• Commit ID: 75940f4
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-cpu
• Commit ID: 74deada
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-win
• Commit ID: 74deada
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-gpu
• Commit ID: 74deada
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-cpu
• Commit ID: 74deada
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-win
• Commit ID: 74deada
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository

[sagemaker-neo-ci-bot]

AWS CodeBuild CI Report

• CodeBuild project: torch-serve-build-gpu
• Commit ID: 74deada
• Result: SUCCEEDED
• Build Logs (available for 30 days)

Powered by github-codebuild-logs, available on the AWS Serverless Application Repository