FedLCM v0.3.0

By downloading, installing or using the software, you accept and agree to be bound by all of the terms and conditions of the LICENSE and DISCLAIMER.

The v0.3 release of FedLCM upgrades its support of FATE to v1.11.1 and OpenFL to v1.5. Highlights of these upgrades includes:

• Support upgrading existing FATE exchange and clusters deployed by previous version of FedLCM.
• Support deploying FATE cluster with GPU support.
• Document update and new image built guide for OpenFL images.
• Support using less Kubernetes privileges to manage OpenFL federations.
• Other bug-fixes & enhancements.

The detailed changes are listed below.

What's Changed

• Add workflows by @owlet42 in #4
• Support FATE 1.10 by @owlet42 in #7
• FedLCM supports upgrading clusters by @owlet42 in #30
• Modify Sample shape check by @qijianshuai in #35
• Supports OpenFL v1.5 by @owlet42 in #34
• fix the cluster not managed by FedLCM can be upgraded by @owlet42 in #37
• Update OpenFL deployment and document to reflect recent changes in v1.5 by @wfangchi in #38
• Update OpenFL guide on image building and example of dummy shard descriptor by @wfangchi in #39
• Fix the problem that the front end modifies the endpoint through yaml by @qijianshuai in #41
• Add subPath to the volumeMount of FedLCM postgres container by @wfangchi in #46
• Support using less privileged kubeconfig to deploy OpenFL components by @wfangchi in #42
• Director sets the namespace according to the selected endpoint by @qijianshuai in #49
• Bind to the data subfolder of the postgres container for FedLCM docker-compose deployment by @wfangchi in #48
• Support FATE v1.11.1 by @wfangchi in #51
• Update swagger API doc by @wfangchi in #52
• Add gpu on the create cluster page by @qijianshuai in #53

Full Changelog: v0.2.0...v0.3.0

FedLCM v0.2.0

By downloading, installing or using the software, you accept and agree to be bound by all of the terms and conditions of the LICENSE and DISCLAIMER.

The v0.2 release of FedLCM contains a Site Portal service that can be deployed via a standalone, docker-compose approach, decoupling it from the Lifecycle Manager service. In addition to that, Site Portal can now start vertical federated learning jobs. And the Lifecycle Manager service now requires less Kubernetes permissions. Other major features and improvements are:

Lifecycle Manager

• Support deploying and managing KubeFATE v1.9, FATE v1.9 and FATE Exchange v1.9 releases.
• Automatically configure the deployed Site Portal and register it to FML Manager. No manual action is needed.
• Support adding Kubernetes infrastructures using namespace-only permissions.
• Support specifying external engines when deploying FATE clusters, and skip deploying these engines as containers.
• When Lifecycle Manager is running in Kubernetes, the underlying Kubernetes can be added automatically as an infrastructure.

Site Portal & FML Manager

• Site Portal and FML Manager can be deployed and running in a standalone fashion using docker-compose. It is still recommended to deploy Site Portal using Lifecycle Manager but it is no longer mandatory.
• When running as a standalone service, Site Portal can work with both FATE with Spark and FATE with Eggroll.
• Support HeteroLR and HeteroSBT type of training and predicting jobs, in addition to the existing HomoLR and HomoSBT jobs.
• Support add existing FATE table into the service for future jobs.
• Support un-registration of the Site Portal from the FML Manager.

The deliverables include:

• The FedLCM Deployment packages for both Docker Compose and Kubernetes
• The Site Portal Docker Compose deployment package
• The FML Manager Docker Compose deployment package
• The FedLCM frontend and server container images
• The Site Portal frontend and server container images
• The FML Manager server container image

FedLCM v0.1.0

By downloading, installing or using the software, you accept and agree to be bound by all of the terms and conditions of the LICENSE and DISCLAIMER.

This is the first release of FedLCM that contains the following components and features:

Lifecycle Manager

• Infrastructure Management: Support adding Kubernetes (any compatible application platforms, e.g. TKG, OpenShift, K3s, KubeEdge, etc.) clusters as infrastructures and configuring FATE registries for each cluster;
• Endpoint Management: Support installing/uninstalling KubeFATE services or add existing KubeFATE services, with support of optionally installing ingress controllers;
• Federation Management: Support creating FATE federations, including configuring the domain names and the deployed participants' route tables, etc.;
• FATE Deployment: Install/Uninstall FATE exchange and clusters on selected Kubernetes clusters; currently support deploying FATE with Spark, HDFS and Pulsar as backend engines;
• Site Portal Deployment: Deploy Site Portal services inside each deployed FATE clusters to enable creating FATE jobs from a Web UI;
• Certificate Management: Work with an embedded or external StepCA service to issue certificates for the newly created FATE components to enable mTLS connections;

Site Portal & FML Manager

• Create federated learning projects and invite other sites to join a project;
• Upload local data into FATE system and associate it into projects;
• Create modeling jobs using drag-n-drop UI or JSON templates; currently support HomoLR and HomoSecureBoost;
• Create predicting jobs using previously trained models;
• Create PSI jobs using associated data;

The deliverables include:

• The FedLCM Deployment packages for both Docker Compose and Kubernetes
• The FedLCM frontend and server container images
• The Site Portal frontend and server container images
• The FML Manager server container image