Research Library

open-access-imgOpen AccessA Survey of Machine Learning for Computer Architecture and Systems
Author(s)
Nan Wu,
Yuan Xie
Publication year2022
Publication title
acm computing surveys
Resource typeJournals
It has been a long time that computer architecture and systems are optimizedfor efficient execution of machine learning (ML) models. Now, it is time toreconsider the relationship between ML and systems, and let ML transform theway that computer architecture and systems are designed. This embraces atwofold meaning: improvement of designers' productivity, and completion of thevirtuous cycle. In this paper, we present a comprehensive review of the workthat applies ML for computer architecture and system design. First, we performa high-level taxonomy by considering the typical role that ML techniques takein architecture/system design, i.e., either for fast predictive modeling or asthe design methodology. Then, we summarize the common problems in computerarchitecture/system design that can be solved by ML techniques, and the typicalML techniques employed to resolve each of them. In addition to emphasis oncomputer architecture in a narrow sense, we adopt the concept that data centerscan be recognized as warehouse-scale computers; sketchy discussions areprovided in adjacent computer systems, such as code generation and compiler; wealso give attention to how ML techniques can aid and transform designautomation. We further provide a future vision of opportunities and potentialdirections, and envision that applying ML for computer architecture and systemswould thrive in the community.
Subject(s)architecture , art , artificial intelligence , compiler , computer architecture , computer science , database centric architecture , programming language , reference architecture , software , software architecture , software engineering , systems architecture , systems design , visual arts
Language(s)English
SCImago Journal Rank2.079
H-Index163
ISSN0360-0300
DOI10.1145/3494523

Seeing content that should not be on Zendy? Contact us.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here