MEEK: Re-thinking Heterogeneous Parallel Error Detection Architecture for Real-World OoO Superscalar Processors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiang, Zhe, Liao, Minli, Ainsworth, Sam, You, Dean, Jones, Timothy
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908296121679872
author Jiang, Zhe
Liao, Minli
Ainsworth, Sam
You, Dean
Jones, Timothy
author_facet Jiang, Zhe
Liao, Minli
Ainsworth, Sam
You, Dean
Jones, Timothy
contents Heterogeneous parallel error detection is an approach to achieving fault-tolerant processors, leveraging multiple power-efficient cores to re-execute software originally run on a high-performance core. Yet, its complex components, gathering data cross-chip from many parts of the core, raise questions of how to build it into commodity cores without heavy design invasion and extensive re-engineering. We build the first full-RTL design, MEEK, into an open-source SoC, from microarchitecture and ISA to the OS and programming model. We identify and solve bottlenecks and bugs overlooked in previous work, and demonstrate that MEEK offers microsecond-level detection capacity with affordable overheads. By trading off architectural functionalities across codesigned hardware-software layers, MEEK features only light changes to a mature out-of-order superscalar core, simple coordinating software layers, and a few lines of operating-system code. The Repo. of MEEK's source code: https://github.com/SEU-ACAL/reproduce-MEEK-DAC-25.
format Preprint
id arxiv_https___arxiv_org_abs_2504_01347
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MEEK: Re-thinking Heterogeneous Parallel Error Detection Architecture for Real-World OoO Superscalar Processors
Jiang, Zhe
Liao, Minli
Ainsworth, Sam
You, Dean
Jones, Timothy
Hardware Architecture
Heterogeneous parallel error detection is an approach to achieving fault-tolerant processors, leveraging multiple power-efficient cores to re-execute software originally run on a high-performance core. Yet, its complex components, gathering data cross-chip from many parts of the core, raise questions of how to build it into commodity cores without heavy design invasion and extensive re-engineering. We build the first full-RTL design, MEEK, into an open-source SoC, from microarchitecture and ISA to the OS and programming model. We identify and solve bottlenecks and bugs overlooked in previous work, and demonstrate that MEEK offers microsecond-level detection capacity with affordable overheads. By trading off architectural functionalities across codesigned hardware-software layers, MEEK features only light changes to a mature out-of-order superscalar core, simple coordinating software layers, and a few lines of operating-system code. The Repo. of MEEK's source code: https://github.com/SEU-ACAL/reproduce-MEEK-DAC-25.
title MEEK: Re-thinking Heterogeneous Parallel Error Detection Architecture for Real-World OoO Superscalar Processors
topic Hardware Architecture
url https://arxiv.org/abs/2504.01347