Purlin: Separating Orchestration from the Datapath of Collectives
Quick summary
arXiv:2609.36954v1 Announce Type: cross Abstract: Distributed inference depends on GPU collective communication that must keep pace with evolving hardware and specialized workloads. However, existing collective implementations often couple semantics, orchestration (where and when data moves), and the datapath (how data moves). This coupling makes it costly to adopt new hardware mechanisms and customize communication for applications. We present Purlin, a scale-up communication framework that separates these concerns. At the top of Purlin, we specify collectives as a naming of an input and outp
Key takeaways
- arXiv:2609.36954v1 Announce Type: cross Abstract: Distributed inference depends on GPU collective communication that must keep pace with evolving hardware and specialized workloads.
- However, existing collective implementations often couple semantics, orchestration (where and when data moves), and the datapath (how data moves).
- This coupling makes it costly to adopt new hardware mechanisms and customize communication for applications.
Why it matters
“Purlin: Separating Orchestration from the Datapath of Collectives” exposes the compute, energy and supply-chain layer behind model competition. Capacity shifts can influence model costs, service availability and the ability of smaller companies to compete.
Member comments