Abstract
Due to imbalances in technology scaling, the energy consumption of data storage and communication by far exceeds the energy consumption of actual data production, i.e., computation. As a consequence, recomputing data can become more energy-efficient than storing and retrieving precomputed data. At the same time, recomputation can relax the pressure on the memory hierarchy and the communication bandwidth. This study hence assesses the energy efficiency prospects of trading computation for communication. We introduce an illustrative proof-of-concept design, identify practical limitations, and provide design guidelines.
| Original language | English (US) |
|---|---|
| Title of host publication | ASPLOS 2017 - 22nd International Conference on Architectural Support for Programming Languages and Operating Systems |
| Publisher | Association for Computing Machinery |
| Pages | 811-824 |
| Number of pages | 14 |
| Volume | 52 |
| Edition | 4 |
| ISBN (Electronic) | 9781450344654 |
| DOIs | |
| State | Published - Apr 4 2017 |
| Event | 22nd International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2017 - Xi'an, China Duration: Apr 8 2017 → Apr 12 2017 |
Publication series
| Name | International Conference on Architectural Support for Programming Languages and Operating Systems - ASPLOS |
|---|---|
| Volume | Part F127193 |
Other
| Other | 22nd International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2017 |
|---|---|
| Country/Territory | China |
| City | Xi'an |
| Period | 4/8/17 → 4/12/17 |
Bibliographical note
Publisher Copyright:© 2017 ACM.
Keywords
- Energy efficiency
- Recomputation
