Skip to content
main
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
src
 
 
 
 
 
 
 
 
 
 
 
 
 
 

ModelMesh

The ModelMesh framework is a mature, general-purpose model serving management/routing layer designed for high-scale, high-density and frequently-changing model use cases. In essence, it acts as a distributed LRU cache for serving runtime models.

If you are looking at deploying and managing ModelMesh, please use the ModelMesh Serving repo instead, as it hosts the controller for the same. The instructions here are for development purposes only.

Quick-Start

  1. Wrap your model-loading and invocation logic in this model-runtime.proto gRPC service interface
    • runtimeStatus() - called only during startup to obtain some basic configuration parameters from the runtime, such as version, capacity, model-loading timeout
    • loadModel() - load the specified model into memory from backing storage, returning when complete
    • modelSize() - determine size (mem usage) of previously-loaded model. If very fast, can be omitted and provided instead in the response from loadModel
    • unloadModel() - unload previously loaded model, returning when complete
    • Use a separate, arbitrary gRPC service interface for model inferencing requests. It can have any number of methods and they are assumed to be idempotent. See predictor.proto for a very simple example.
    • The methods of your custom applier interface will be called only for already fully-loaded models.
  2. Build a grpc server docker container which exposes these interfaces on localhost port 8085 or via a mounted unix domain socket
  3. Extend the Kustomize-based Kube configs to use your docker image, and with appropriate mem and cpu resource allocations for your container
  4. Deploy to a Kubernetes cluster as a regular Service, which will expose this grpc service interface via kube-dns (you do not implement this yourself), consume using grpc client of your choice from your upstream service components
    • registerModel() and unregisterModel() for registering/removing models managed by the cluster
    • Any custom inferencing interface methods to make a runtime invocation of previously-registered model, making sure to set a mm-model-id or mm-vmodel-id metadata header (or -bin suffix equivalents for UTF-8 ids)

Deployment and Upgrades

Prerequisites:

  • An etcd cluster (shared or otherwise)
  • A Kubernetes namespace with the etcd cluster connection details configured as a secret key in this json format
    • Note that if provided, the root_prefix attribute is used as a key prefix for all of the framework's use of etcd

From an operational standpoint, ModelMesh behaves just like any other homogeneous clustered microservice. This means it can be deployed, scaled, migrated and upgraded as a regular Kubernetes deployment without any special coordination needed, and without any impact to live service usage.

In particular the procedure for live upgrading either the framework container or service runtime container is the same: change the image version in the deployment config yaml and then update it kubectl apply -f model-mesh-deploy.yaml

Build

Sample build:

GIT_COMMIT=$(git rev-parse HEAD)
BUILD_ID=$(date '+%Y%m%d')-$(git rev-parse HEAD | cut -c -5)
IMAGE_TAG_VERSION=0.0.1
IMAGE_TAG=${IMAGE_TAG_VERSION}-$(git branch --show-current)_${BUILD_ID}

docker build -t model-mesh:${IMAGE_TAG} \
    --build-arg imageVersion=${IMAGE_TAG} \
    --build-arg buildId=${BUILD_ID} \
    --build-arg commitSha=${GIT_COMMIT} .

About

Distributed Model Serving Framework

Resources

License

Packages

No packages published

Languages