Ripple: Asynchronous Programming for Spatial Dataflow Architectures

Spatial dataflow architectures (SDAs) are a promising and versatile accelerator platform. They are software-programmable and achieve near-ASIC performance and energy efficiency, beating CPUs by orders of magnitude. Unfortunately, many SDAs struggle to efficiently implement irregular computations bec...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of ACM on programming languages Ročník 9; číslo PLDI; s. 249 - 276
Hlavní autoři: Ghosh, Souradip, Shi, Yufei, Lucia, Brandon, Beckmann, Nathan
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York, NY, USA ACM 10.06.2025
Témata:
Computer systems organization
Data flow architectures
Parallel architectures
Software and its engineering
spatial dataflow architectures
asynchronous programming
ISSN:2475-1421, 2475-1421
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Spatial dataflow architectures (SDAs) are a promising and versatile accelerator platform. They are software-programmable and achieve near-ASIC performance and energy efficiency, beating CPUs by orders of magnitude. Unfortunately, many SDAs struggle to efficiently implement irregular computations because they suffer from an abstraction inversion: they fail to capture coarse-grain dataflow semantics in the application — namely asynchronous communication, pipelining, and queueing — that are naturally supported by the dataflow execution model and existing SDA hardware. Ripple is a language and architecture that corrects the abstraction inversion by preserving dataflow semantics down the stack. Ripple provides asynchronous iterators, shared-memory atomics, and a familiar task-parallel interface to concisely express the asynchronous pipeline parallelism enabled by an SDA. Ripple efficiently implements deadlock-free, asynchronous task communication by exposing hardware token queues in its ISA. Across nine important workloads, compared to a recent ordered-dataflow SDA, Ripple shrinks programs by 1.9×, improves performance by 3×, increases IPC by 58%, and reduces dynamic instructions by 44%.
ISSN:2475-1421
2475-1421
DOI:10.1145/3729256