OpenPAI
MicrosoftCluster management and resource scheduling platform for AI, originally from Microsoft Research. Users submit deep-learning jobs to a shared GPU cluster; OpenPAI schedules them with HiveD, a scheduler extender that models each tenant as a virtual cluster and guarantees not just a GPU count but the topology those GPUs sit in, so an eight-GPU job that must run inside one node gets a placement that satisfies it.
The HiveD scheduler is vendored at src/hivedscheduler and also published separately as microsoft/hivedscheduler.
Openness
5 high confidence- license
- MIT(OSI
- source
- public(microsoft/pai carries the platform and vendors its scheduler at src/hivedscheduler)
- core features withheld
- no — a 2,498-entry recursive tree carries no path matching enterprise, ee, commercial or proprietary
The OpenPAI LICENSE is the standard MIT grant under a Microsoft Corporation copyright line, read from the body rather than the GitHub label. A full scan of the repository tree finds no enterprise, ee or commercial path, and Microsoft sells no paid edition, so nothing is withheld from what is published.
- https://api.github.com/repos/microsoft/pai/git/trees/master?recursive=1 recorded 2026-09-15
Full untruncated recursive tree of the default branch, 2,498 entries. Licenses are the root LICENSE and vendored dependencies. No path matches enterprise, ee, commercial or proprietary. A tree lists paths; it is cited for that and for nothing about the vendor's offerings.
- https://github.com/microsoft/pai recorded 2026-09-15
The microsoft/pai repository page. Microsoft publishes no commercial OpenPAI edition, and the repository is the whole platform; this is cited for that and for the license.
- https://raw.githubusercontent.com/microsoft/pai/master/LICENSE recorded 2026-09-17
The LICENSE body, read rather than the classifier label: the verbatim MIT text under "Copyright (c) Microsoft Corporation. All rights reserved.", indented four spaces throughout.
Adoption
2 low confidenceOpenPAI ships as container images and a Helm chart rather than through a package registry, leaving GitHub stars as the only adoption signal. A star is not a use, so the signal shows attention rather than real usage.
- https://github.com/microsoft/pai recorded 2026-09-15
Repository page for microsoft/pai showing 2,690 stars.
Capability
4 medium confidenceHiveD decides which GPUs on which machines a job receives and guarantees the topology of that set, not just its size, so an eight-GPU job that must run inside one node is placed only where that holds. Its manual describes cell types representing topology above the node level, but that is a description rather than evidence that one job is ever granted a reserved multi-machine set.
- https://github.com/microsoft/pai recorded 2026-09-15
Repository page for microsoft/pai, establishing that the platform being scored is the one HiveD schedules for.
- https://raw.githubusercontent.com/microsoft/hivedscheduler/master/doc/user-manual.md recorded 2026-09-15
HiveD's user manual: a cellType 'defines a resource topology of a skuType', with skuTypes as leaf cellTypes that 'do not have internal topology anymore' and larger cellTypes composed above them. Establishes that the guaranteed unit is a topology cell rather than a GPU count, and that cells exist above the single-node level.
- https://raw.githubusercontent.com/microsoft/hivedscheduler/master/README.md recorded 2026-09-15
HiveD's own README: 'HiveD is a scheduler for deep learning workloads', a Kubernetes scheduler extender for multi-tenant GPU clusters, whose 'killer feature' is topology-aware resource guarantee - the example given is an eight-GPU job that must run within a single node and cannot be allocated by a scheduler that guarantees only a count.
Verified 2026-09-15