Trireme: Exploring Hierarchical Multi-Level Parallelism for Domain Specific Hardware Acceleration

01/21/2022
by   Georgios Zacharopoulos, et al.
0

The design of heterogeneous systems that include domain specific accelerators is a challenging and time-consuming process. While taking into account area constraints, designers must decide which parts of an application to accelerate in hardware and which to leave in software. Moreover, applications in domains such as Extended Reality (XR) offer opportunities for various forms of parallel execution, including loop level, task level and pipeline parallelism. To assist the design process and expose every possible level of parallelism, we present Trireme, a fully automated tool-chain that explores multiple levels of parallelism and produces domain specific accelerator designs and configurations that maximize performance, given an area budget. Experiments on demanding benchmarks from the XR domain revealed a speedup of up to 20x, as well as a speedup of up to 37x for smaller applications, compared to software-only implementations.

READ FULL TEXT

page 5

page 8

research
07/02/2019

Accelerator-level Parallelism

Future applications demand more performance, but technology advances hav...
research
07/18/2020

Design Space Exploration of Algorithmic Multi-Port Memories in High-Performance Application-Specific Accelerators

Memory load/store instructions consume an important part in execution ti...
research
05/31/2023

ReDSEa: Automated Acceleration of Triangular Solver on Supercloud Heterogeneous Systems

When utilized effectively, Supercloud heterogeneous systems have the pot...
research
09/16/2023

Rewriting History: Repurposing Domain-Specific CGRAs

Coarse-grained reconfigurable arrays (CGRAs) are domain-specific devices...
research
03/06/2023

Domain-Specific Computational Storage for Serverless Computing

While (1) serverless computing is emerging as a popular form of cloud ex...
research
10/20/2021

Synthesizing Optimal Parallelism Placement and Reduction Strategies on Hierarchical Systems for Deep Learning

We present a novel characterization of the mapping of multiple paralleli...
research
04/05/2021

Meta-level issues in Offloading: Scoping, Composition, Development, and their Automation

This paper argues for an accelerator development toolchain that takes in...

Please sign up or login with your details

Forgot password? Click here to reset