v0.4.1

Summary

System is more stable

• Fixed bugs that worked as bottleneck to the system such as #608. It is much more stable to be used in the production environment.
• Updated deprecated dependencies that might have potential security issues.

API Update

• Keras deployer is supported now.
• Several API updates including delete_versioned_mode and delete_model_links (#603). (API doc will be updated shortly)
• query_frontend http configuration is exposed to query_frontend

Kubernetes Support Update

• Multiple clusters can be running side by side on the same K8s cluster (Multi tenancy support)
• Prometheus RBAC is supported as default, so that you don't need to create the one for Kubernetes setup.
• K8s service type is configurable. It supports such as LoadBalancer. (API doc will be updated shortly)

Changelog

New Features:

• Multi tenancy support (#503)
• Change the RPCServicePort constant to the global configuration (#660)
• Implement delete_model_links API (#654)
• Expose http configurations to clipper_admin (#524)
• Fluentd Logging System Part 1 (#652)
• Rename 'delete_application' to 'unregister_application' in clipper_admin (#653)
• Keras deployer (#594)
• Change k8s ServiceTypes to be changeable (#667)
• Implement delete_versioned_model API (#602)
• #692: Added get_metric_addr to the ClipperConnection (#693)
• Adding RBAC support in kubernetes config (V2) (#694)

Fix & Dependency Update:

• Backport 0.3.0 fixes (#511)
• Fix docker version to be greater than 3.0 (#515)
• Fix 100% CPU utilization issue [V2] (#608)
• Upgrade requirements and pyyaml versions for security issues (#647)
• Fix a bug that ManagementFrontend adds a garbage to http response (#582)
• Fix Python anti pattern (#651)
• change set_num_replicas in docker_container_manager (#616)
• Revert ClipperLibBaseDockerfile (#622)
• Specify PyTorch version in use (#664)
• Use CLIPPER_REGISTRY, CLIPPER_TAG environment variable if existed (#666)
• add urllib3 version restriction to requirements.txt (#679)
• Remove metric log from query frontend (#680)
• Enable TCP keep alive for redox (#717)
• Added a return statement (#697)
• Fix a bug that a new model replica couldn't be registerd with the Clipper's RPCService (#689)
• Updagrade the version of mxnet to 1.4.* for mxnetcontainerdockerfile. (#675)
• Upgrade Tensorflow version(1.6 -> 1.13) (#720)
• Handle trailing slash on predict url (#721)
• Update SimpleWebServer to the latest version(29 Apr, 2019 - d6985ed6) (#691)
• Print necessary stackstace information when crashed (#700)
• Clean up the versioned model's batch_size and metrics when deleting it (#711)
• Update the version of base libraries and OS (#714)
• Add check for external redis ip address to prevent KeyError (#571) (#719)

Infra:

• [CI] Several CI/Infrastructure Improvement & Fix (#595)
• [CI] Cleaning up #595 (#609)
• Try to fix CI (#649)
• [CI] Remove the dependency about Debian jessie backport (#658)
• [CI] Update the apt package list before installing openjdk (#659)
• [CI] Add retry-routine to 'docker push' in clipper_docker.cfg.py (#655)
• Decrease test samples for test_stop_models case (#657)
• Increase sleep time from 500us to 1000us in timers_test to avoid unintended errors (#661)
• Retry when docker push fails (#662)
• Fix some bugs of PySpark integration test & Disable Spark UI (#665)
• Catch more errors when pushing a Docker image (#663)
• [CI] Try to stabilize Travis job (#685)
• [CI] Randomize RPCService's port and handle NoSuchProcess error properly (#690)
• [CI] Recover CI job failures (#699)
• [CI] Fix maintenance scripts (#707)

Doc:

• Update Basic example README.md (#600)
• Fix typo (#586)
• Example (#644)

v0.3.0

commit • Only link models to apps in k8s test the first time model is registered (#290)

commit • update benchmark readme (#293)

commit • Batch size estimation fix (#294)

commit • R Model Deployment (#261)

commit • Minor cleanup to rpc.py (#295)

commit • Update docker publish script to publish minor version tags (#298)

commit • Fix typos in k8s method calls. (#299)

commit • Fix broken links (#303)

commit • Add CLIPPER_INPUT_TYPE to environment when launching k8s model container. (#304)

commit • Fix Issue with ClipperException and Docker not running case (#308)

commit • Support InternalIP address for K8S (#316)

commit • Support for Generic Tensorflow Models (#319)

commit • Converted path to posixpath in python deployer (#318)

commit • Fix a broken link in the README.md (#323)

commit • Fixed conda object manipulation problem (#297)

commit • Support Deployment of SparkML Models (#324)

commit • Handle exceptions associated with loading nonexistent libraries (#344)

commit • Support deploying Saved Tensorflow Models (#328)

commit • Batch predict (#321)

commit • Support for PyTorch model (#346)

commit • Adding function: Allow users to designate the number of batches (#320)

commit • Metrics Monitoring: A minimal viable product (#348)

commit • Fix dict unpacking lint error (#356)

commit • Support Boost 1.66.0 (#358)

commit • Turn off ssl verification to fix conda http error (#369)

commit • Fix error (#373)

commit • Fix port already bound error in integration tests (#368)

commit • Metrics and Monitoring V0.0.2: Model Container, Latencies, and Example (#357)

commit • Integration test for fixed-batch-size model deployment (#370)

commit • fixed decoding error when when parsing graph_def from string. (#377)

commit • Metric Monitoring: Set Prometheus Version Tag (#384)

commit • replaced queues in RPC layer with MoodyCamel's lock-free queues (#380)

commit • Handle ChunkedEncodingError on check_solvability_write_deps() (#398)

commit • Support latest conda version (#385)

commit • Metric Monitoring: Add to/Delete from the metric config of Prometheus (#402)

commit • Fix flakiness in fixed batch size integration test (#387)

commit • StateDB Cache (#350)

commit • Explicitly install redis-server when building test container (#410)

commit • Added MXNet Deployment Capability (replaces #390) (#405)

commit • [Metric] Refactor Frontend-exporter (#413)

commit • Improved README with direct links to github profiles (#414)

commit • Fix a bug that QueryFrontend adds a garbage to http response (#415)

commit • PyTorch + ONNX + Caffe2 Model deployer (#362)

commit • results of profiling JSON parsing latency (#418)

commit • Fix installation order errors in deployer containers (#426)

commit • Container Manager Wait Until Ready (#417)

commit • fix libsodium issue (#432)

commit • [Metrics] Frontend-exporter fault tolerance quick fix (#429)

commit • Model container testing function to clipper admin (Vanilla python) (#394)

commit • Upgrade SimpleWebServer to version 2.1.1 (#438)

commit • [Metrics] Kubernetes Support & Maintenance (#427)

commit • Add ability to install packages + xgboost model deployer + example (#431)

commit • Add missing integration tests, clean up dockerfiles, and remove Anaconda dependencies (#435)

commit • Implement Metric Server (#440)

commit • Fix QueryFrontend bug (#458)

commit • Ensure container/model is ready to be queried (#447)

commit • Fix hard coded checkpoint prefix for tensorflow deployer. (#464)

commit • Add functionality to delete applications (#457)...

v0.2.0

Release Notes

Major Changes:

• Major refactor of the Clipper admin Python library to make it more robust and modular
• Support for running Clipper on Kubernetes through the KubernetesContainerManager

Changelog

commit • minor fix to unittest script for jenkins (#202)

commit • Fix example code in README (#205)

commit • removed pip install findspark from run_unittests.sh (#211)

commit • fixed backslash escape issue for removing remote containers (#210)

commit • Restarting on containers is no longer the default behavior (#213)

commit • Base64 decoding for JSON byte data (#214)

commit • Allow model versions to be strings (#197)

commit • R model deployment (#195)

commit • One-line app registration and model deployment (#223)

commit • [CLIPPER-227] Fix EWMA behavior for meters (#228)

commit • RPC container content fix (#232)

commit • minor fix to unittest script for jenkins (#202)

commit • Fix example code in README (#205)

commit • removed pip install findspark from run_unittests.sh (#211)

commit • fixed backslash escape issue for removing remote containers (#210)

commit • Restarting on containers is no longer the default behavior (#213)

commit • Base64 decoding for JSON byte data (#214)

commit • Allow model versions to be strings (#197)

commit • R model deployment (#195)

commit • One-line app registration and model deployment (#223)

commit • [CLIPPER-227] Fix EWMA behavior for meters (#228)

commit • RPC container content fix (#232)

commit • Wording fix (#234)

commit • Fixed broken R container (#247)

commit • Handling for internal server errors (#246)

commit • Update Python container to correctly run "pip install cryptography" (#243)

commit • Updated libmzmq3 to libzmq5 (#248)

commit • Benchmarking utilities (#225)

commit • App/model links (add only) (#226)

commit • Make mgmt frontend check for input type object equality (#250)

commit • Save module dependencies not in conda/pip (#237)

commit • Pass by value where appropriate (#252)

commit • C++ Model Container (#249)

commit • Updated README to clarify quickstart (#257)

commit • Removed redis cli in favor of clipper.get_model_info (#260)

commit • Fix docker image versioning (#264)

commit • [CLIPPER-224] Replace boost futures with folly futures (#266)

commit • Management frontend: Improve clarity of response messages (#271)

commit • Fix histogram overflow errors (#272)

commit • [CLIPPER-212] Add CLOCK cache eviction policy (#268)

commit • change CODE_VERSION build arg to develop (#273)

commit • Make unit tests run in a Docker container (#274)

commit • Refactor and clean up Clipper admin python package (#251)

commit • remove default label (#279)

commit • Installs latest version of pyspark via pip install (#282)

commit • Fixes readthedocs documentation (#278)

commit • Build new docker images in build_model with in-memory Dockerfile (#280)

commit • Update docker build script to optionally publish images to DockerHub (#283)

commit • Pick up kube credentials from env variables for k8s integration test (#284)

commit • Include k8s yaml files in pip package (#286)

v0.1.4

This release added three new features as well as number of bugfixes:

• Deploy R models directly from a Python/R training script
• Deploy a model and register the corresponding application in a single method
• Allow model versions to be arbitrary strings. Model versions can now be git hashes, semantic versions, etc. along with integers.

0.1.3

• Fixes several minor bugs and usability issues.
• It also adds support for automatic deployment of PySpark models through clipper_admin.Clipper.deploy_pyspark_model.