A stream compiler for communication-exposed architectures

Abstract

With the increasing miniaturization of transistors, wire delays are becoming a dominant factor in microprocessor performance. To address this issue, a number of emerging architectures contain replicated processing units with software-exposed communication between one unit and another (e.g., Raw, SmartMemories, TRIPS). However, for their use to be widespread, it will be necessary to develop compiler technology that enables a portable, high-level language to execute efficiently across a range of wire-exposed architectures.In this paper, we describe our compiler for StreamIt: a high-level, architecture-independent language for streaming applications. We focus on our backend for the Raw processor. Though StreamIt exposes the parallelism and communication patterns of stream programs, some analysis is needed to adapt a stream program to a software-exposed processor. We describe a partitioning algorithm that employs fission and fusion transformations to adjust the granularity of a stream graph, a layout algorithm that maps a stream graph to a given network topology, and a scheduling strategy that generates a fine-grained static communication pattern for each computational element.We have implemented a fully functional compiler that parallelizes StreamIt applications for Raw, including several load-balancing transformations. Using the cycle-accurate Raw simulator, we demonstrate that the StreamIt compiler can automatically map a high-level stream abstraction to Raw without losing performance. We consider this work to be a first step towards a portable programming model for communication-exposed architectures.

Distributed S-Net: Cluster and Grid Computing without the Hassle

“…S-NET shares the underlying concept of stream processing with a number of synchronous streaming languages, such as Esterel [18] or StreamIt [19]. Asynchronous computing and the separation of concerns between computing and coordination set S-NET apart from those approaches.…”

Streaming Data Movement for Real-Time Image Analysis

“…Stream processors operate on sequences of data using localized computation kernels [16,17,18]. They have the inherent ability to hide latency, amortize instruction overhead, and expose data parallelism by operating on large sets of data.…”

Language and Compiler Design for Streaming Applications

“…(19) The principal technique for doing this involves fusing adjacent filters in the stream graph to form a single filter. Vertical fusion performs fusion on successive filters in a pipeline, while horizontal fusion joins the parallel streams in a split-join.…”