OneSciencePlace (OSP) is a composable platform in active deployment across institutional and national-scale projects. This roadmap outlines the platform areas we are developing next: expanding capabilities, deeper integrations, and open-source community distribution.

OSP Today

OSP is currently available as a fully managed service with active deployments underway covering:

  • Browser-based delivery of interactive web, graphical, and batch applications
  • No-code application UI builder
  • Compute integration with Slurm-based HPC clusters, standalone Linux hosts, and national resources, including ACCESS and NAIRR
  • POSIX and S3-compatible object storage
  • Data sharing with metadata, annotation, and granular access controls
  • FAIR publishing with DOIs, configurable metadata schemas, and multistage curation workflows
  • Federated identity via CILogon, Globus Auth, LDAP, and SAML
  • Role-based access control (RBAC)
  • Content management and site building via Drupal

OSP is being deployed across a growing range of institutional contexts — science gateways, campus HPC portals, national-scale resources, lab environments, and instructional computing.

Explore current platform capabilities

Next Steps - Planned

The next phase of OneSciencePlace is a more compelling research and education platform.

Collaborative Groups — In Development: Shared environments for research teams, classes, and projects to organize resources, manage members, and publish results together. Group spaces will support delegated administration, group-level access controls, and policy-based permissions across all platform capabilities.

Learning Management Integration — Planned: Integration with Canvas LMS, enabling instructors to deploy advanced computational environments for coursework, beyond the Jupyter ecosystem. Course rosters map directly to OSP groups, enabling scalable instructional computing on-campus, cloud, or HPC resources.

Dashboard refinement — In Development: A single administrative and researcher-facing dashboard for managing projects, applications, data, jobs, workflows, and publications — designed for clarity, low cognitive load, and productivity across all OSP capabilities.

FAIR Data Publishing — Planned: A technically rigorous implementation of FAIR principles, drawing on fairsharing.org as a model for standardized metadata schemas, controlled vocabularies, and interoperability standards. FAIR implementation will be discipline-aware, providing a common extensible foundation while supporting domain-specific metadata schemas, vocabularies, and curation workflows configured per deployment.

Globus integration— Planned: Expanded integration with Globus for high-performance data transfer between storage systems, compute resources, and user endpoints. This will complement OSP's existing Globus Auth support for federated identity, adding reliable large-scale data movement across institutional and national resources directly within the platform.

Kubernetes and AI/ML Workloads — Prototyped: Expanded compute integration supporting Kubernetes-based job execution, interactive GPU sessions, and AI/ML workflows. This will enable OSP deployments to support modern machine-learning and data-intensive workloads alongside traditional HPC batch jobs — within the same unified interface.

Workflow Orchestration — Planned: A visual, event-driven workflow composer enabling researchers to build and run multi-step, multi-application pipelines across heterogeneous compute systems. Workflows will be reusable, versioned, and integrated with OSP's provenance and publishing infrastructure — enabling end-to-end reproducible research directly from the platform.

Post-Run Job Archiving — Planned: Option to archive job outputs to POSIX or S3 storage after execution completes.

External API Layer — Planned: A programmable API layer enabling automation, interoperability, and integration with external tools and services. This will allow institutions and developers to build on OSP's capabilities programmatically rather than exclusively through the web interface.

Open-Source Community Distribution — Planned A structured public release of the unified OSP distribution, enabling institutions to self-host, contribute, and adapt the platform to their specific infrastructure and community needs.

Learn more about the open-source roadmap

Interested in contributing to OSP's development or discussing the roadmap? Contact the OSP team