Livia: Data-Centric Computing Throughout the Memory Hierarchy
Name
3373376.3378497.pdf
Size
1.57 MB
Format
Adobe PDF
Checksum (MD5)
ed59b352bc58fb1d7997ab6bfec0bc56
Author(s)
Lockerman, Elliot
Feldmann, Axel
Bakhshalipour, Mohammad
Stanescu, Alexandru
Gupta, Shashwat
Sanchez, Daniel
Beckmann, Nathan
Date Issued
2020
Journal
International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS
Publisher
Association for Computing Machinery (ACM)
Citation
Lockerman, Elliot, Feldmann, Axel, Bakhshalipour, Mohammad, Stanescu, Alexandru, Gupta, Shashwat et al. 2020. "Livia: Data-Centric Computing Throughout the Memory Hierarchy." International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS.
Version
Final published version
Abstract
© 2020 Copyright held by the owner/author(s). Publication rights licensed to ACM. In order to scale, future systems will need to dramatically reduce data movement. Data movement is expensive in current designs because (i) traditional memory hierarchies force computation to happen unnecessarily far away from data and (ii) processing-in-memory approaches fail to exploit locality. We propose Memory Services, a flexible programming model that enables data-centric computing throughout the memory hierarchy. In Memory Services, applications express functionality as graphs of simple tasks, each task indicating the data it operates on. We design and evaluate Livia, a new system architecture for Memory Services that dynamically schedules tasks and data at the location in the memory hierarchy that minimizes overall data movement. Livia adds less than 3% area overhead to a tiled multicore and accelerates challenging irregular workloads by 1.3× to 2.4× while reducing dynamic energy by 1.2× to 4.7×.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3373376.3378497
