OKDP logoOKDP

OKDP Roadmap

Track the content, priorities and progress by version.

Complete — delivered, tested and usable
In progress — active development or integration
To do — identified, but not started for this version

v1.0.0

September 2026

Data & AI Modules

A catalog of reference open source tools. Usable individually or combined, without depending on the OKDP Control Plane (UI/Server).

Lakehouse & Analytics

Apache Spark
✓
  • ✓Dedicated Helm chart for Spark History Server
  • ✓Dedicated authentication plugin (Spark Auth Proxy)
  • ✓Spark Web Proxy extension (real-time job UI access)
  • ✓Spark Docker images with automated build pipeline
Trino
✓
  • ✓Community Images & Helm Charts
  • ✓OIDC Authentication
  • ✓Polaris Catalog integration (Lakehouse connectivity)
  • ✓Fine-grained authorization via OPA (Open Policy Agent)
Polaris Catalog (Apache Iceberg)
✓
  • ✓Community Images & Helm Charts
  • ✓Full integration with Trino
  • ✓Native OIDC authentication
  • ✓S3 Integration & Connectivity (STS)
  • ✓Identity propagation from Trino to Polaris
  • ✓Fine-grained OPA integration (Open Policy Agent)
  • ✓Make STS support optional (infrastructure flexibility)

Visualization & BI

Apache Superset
✓
  • ✓Community Images & Helm Charts
  • ✓OIDC Authentication

Orchestration & Governance

Apache Airflow
✓
  • ✓Community Images & Helm Charts
  • ✓OIDC Authentication

Data Science

JupyterHub / JupyterLab
✓
  • ✓Community Images & Helm Charts
  • ✓Helm Values tuning
  • ✓OKDP-optimized Docker images for JupyterLab environments
  • ✓Community images (Hub, Proxy, Culler)

Control Plane / Platform

OKDP's automation layer: orchestration, multi-tenant isolation, and developer experience (DX) for a turnkey experience across the entire stack.

Admin Console

↻
  • ✓Global project management (UI/Backend)
  • ✓Target Kubernetes cluster configuration
  • ✓Local users and groups management (via Kubauth)
  • ↻Project membership management

Project Console

↻
  • ✓Isolated spaces creation via Project CRD (basic RBAC)
  • ✓JupyterHub environment provisioning
  • ✓Spark History Server instance deployment
  • ✓Project Secrets management (Vault via token only)
  • ✓Application submission via Spark Operator (UI and kubectl)
  • ↻Running jobs listing and monitoring
  • ↻Control Plane integration: Trino, Superset, Polaris and Airflow

Sandbox

↻
  • ✓Local sandbox (Kind + Flux) with legacy components
  • ✓Sandbox Object Storage (SeaweedFS)
  • ↻Ingress / API Gateway migration
  • ↻New okdp-server & okdp-ui versions
  • ○New okdp-operator component

Installation & Guides

↻
  • ✓Demo Application: End-to-end use case (Ingestion, Spark & Airflow)
  • ↻Control Plane Helm chart (okdp-platform)
  • ○Admin Guide: Platform Deployment & Configuration
  • ○User Guide: Module & service usage via console

Future Evolutions

Themes and modules identified to enrich the platform after the v1.0.0 release.

Data & AI Modules

AI & MLOps

↻
  • ↻Kubeflow — Initial integration (Core components)
  • ○MLflow — Lifecycle & Experiment Tracking
  • ○LLM Serving — Language model inference (vLLM, TGI)

Ingestion, Streaming & Governance

○
  • ○Apache NiFi — Connectivity & ETL Ingestion
  • ○Apache Kafka — Messaging & Event Streaming
  • ○Apache Flink — Real-time Processing
  • ○OpenMetadata — Discovery & Global Cataloging

Control Plane / Platform

Platform

○
  • ○Platform RBAC — Fine-grained permission management
  • ○Generic Service Catalog — Self-service interface for deploying custom extra services
  • ○Quota Management — Fine-grained Kubernetes resource management
  • ○Marketplace — Expose custom modules or components

Observability & Deployment

○
  • ○Platform Health — Centralized module metrics & logs collection
  • ○Full GitOps Mode — Support for full GitOps deployment mode