Fast Shared-Memory Barrier Synchronization for a 1024-Cores RISC-V Many-Core Cluster

TL;DR

This paper develops and benchmarks optimized shared-memory barrier synchronization methods for a 1024-core RISC-V cluster, significantly reducing synchronization overhead in high-performance parallel applications.

Contribution

It introduces the first shared-memory barrier implementation for a 1024-core RISC-V cluster and demonstrates substantial performance improvements over naive methods.

Findings

Less than 10% synchronization overhead in benchmarks

1.6x speed-up with optimized tree barriers

6.2% overhead in a 5G application

Abstract

Synchronization is likely the most critical performance killer in shared-memory parallel programs. With the rise of multi-core and many-core processors, the relative impact on performance and energy overhead of synchronization is bound to grow. This paper focuses on barrier synchronization for TeraPool, a cluster of 1024 RISC-V processors with non-uniform memory access to a tightly coupled 4MB shared L1 data memory. We compare the synchronization strategies available in other multi-core and many-core clusters to identify the optimal native barrier kernel for TeraPool. We benchmark a set of optimized barrier implementations and evaluate their performance in the framework of the widespread fork-join Open-MP style programming model. We test parallel kernels from the signal-processing and telecommunications domain, achieving less than 10% synchronization overhead over the total runtime for problems that fit TeraPool's L1 memory. By fine-tuning our tree barriers, we achieve 1.6x speed-up with respect to a naive central counter barrier and just 6.2% overhead on a typical 5G application, including a challenging multistage synchronization kernel. To our knowledge, this is the first work where shared-memory barriers are used for the synchronization of a thousand processing elements tightly coupled to shared data memory.