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


Loading...

Date

2023

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric

Data

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 4 MB 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.

Publication status

published

Book title

Embedded Computer Systems: Architectures, Modeling, and Simulation

Volume

14385

Pages / Article No.

241 - 254

Publisher

Springer

Event

23rd International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS 2023)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Many-core; RISC-V; Synchronization; 5G

Organisational unit

03996 - Benini, Luca / Benini, Luca check_circle

Notes

Funding

Related publications and datasets