一句话：提出语言与架构 Ripple，用异步迭代器、共享内存原子操作与任务并行接口消除抽象倒置，支持非规则计算。
空间数据流架构（SDA）是可编程、接近 ASIC 能效的加速器平台，但在实现非规则计算时存在抽象倒置：应用中的粗粒度数据流语义（异步通信、流水线、队列）本可被数据流执行模型与现有 SDA 硬件自然支持，却难以被高效表达。Ripple 是一种在整栈中保留数据流语义的语言与架构，提供异步迭代器、共享内存原子操作以及熟悉的任务并行接口，以简洁地表达 SDA 所支持的异步流水线并行；其 ISA 暴露硬件 token 队列，从而高效实现无死锁的异步任务通信。在九个重要工作负载上，与近期有序数据流 SDA 相比，Ripple 使程序规模缩小 1.9 倍，性能提升 3 倍，IPC 提高 58%，动态指令数减少 44%。
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%.
@article{paperbot3133,
title = {Ripple: Asynchronous Programming for Spatial Dataflow Architectures},
author = {Souradip Ghosh and Yufei Shi and Brandon Lucia and Nathan Beckmann},
journal = {Proceedings of the ACM on Programming Languages},
volume = {9},
number = {PLDI},
year = {2025},
doi = {10.1145/3729256}
}