triton-inference-server / triton-inference-server/local_cache

Local cache on sagemaker

Open
#11 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
6
Forks
6
Avg merge
14h 24m
Merged PRs (30d)
2

Description

Experiencing this error when starting a sagemaker endpoint with local-cache:
error: creating server: Invalid argument - unable to find 'libtritoncache_local.so' for cache. Searched: /opt/tritonserver/caches

endpoint image: 746233611703.dkr.ecr.eu-central-1.amazonaws.com/sagemaker-tritonserver:23.12-py3
environment variables:

"Environment": {
                "SAGEMAKER_TRITON_DEFAULT_MODEL_NAME": "ensemble",
                "SAGEMAKER_TRITON_THREAD_COUNT": "10",
                "SAGEMAKER_TRITON_LOG_INFO": "true",
                "SAGEMAKER_TRITON_LOG_VERBOSE": "true",
                
                "SAGEMAKER_TRITON_ALLOW_METRICS": "true",
                "SAGEMAKER_TRITON_PUBLISH_METRICS_TO_CLOUDWATCH": "true",
                "SAGEMAKER_TRITON_CLOUDWATCH_LOG_GROUP": f"/aws/SageMaker/Endpoints/TritonMetrics/{EXPERIMENT}",
                "SAGEMAKER_TRITON_CLOUDWATCH_METRIC_NAMESPACE": f"/aws/SageMaker/Endpoints/TritonMetrics/{EXPERIMENT}",
                "SAGEMAKER_TRITON_METRICS_PORT": "22000",
                "SAGEMAKER_TRITON_ADDITIONAL_ARGS": "--cache-config local,size=1000000",
            },

Any idea ho to add libtritoncache_local to sagemaker image ?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the endpoint startup with the sagemaker-tritonserver:23.12-py3 image and --cache-config local,size=1000000. Check how this repository builds and packages libtritoncache_local.so, then compare that output with the SageMaker image's /opt/tritonserver/caches path. Done means the endpoint starts with local caching enabled without the missing-library error.

Written by the indexing model from the issue text.

Assessment

Tech stack
aws, cpp
Domain
cloud
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.