WRSWorld Research StateContribute
WORLD RESEARCH STATE

Groq TSP: compile-time scheduled tensor-streaming architecture

Contribution grade

Deterministically scheduled AI accelerators · Human-designed research method or resource

Reference score150,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Built a tensor-streaming processor around functional slices, producer-consumer streams and compiler-controlled cycle-level scheduling, explicitly managing computation and SRAM data movement.

Scientific scope

Built a tensor-streaming processor around functional slices, producer-consumer streams and compiler-controlled cycle-level scheduling, explicitly managing computation and SRAM data movement. Static scheduling/dataflow and tensor acceleration have predecessors. Determinism pertains to the compiled model, instruction/data schedule and stated hardware conditions, not arbitrary external services or all workloads. Original2020 ISCA implementation only; inaccessible2022 scale-out extension is not silently included. Source is an author-uploaded reproduction, not a new2022 publication or independently reproduced benchmark.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Static scheduling/dataflow and tensor acceleration have predecessors. Determinism pertains to the compiled model, instruction/data schedule and stated hardware conditions, not arbitrary external services or all workloads. Original2020 ISCA implementation only; inaccessible2022 scale-out extension is not silently included. Source is an author-uploaded reproduction, not a new2022 publication or independently reproduced benchmark.

Contributors

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction