Directoryless shared memory coherence using execution migration
We introduce the concept of deadlock-free migration-based coherent shared memory to the NUCA family of architectures. Migration-based architectures move threads among cores to guarantee sequential semantics in large multicores. Using a execution migration (EM) architecture, we achieve performance co...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
ACTA Press
2012
|
| Online Access: | http://hdl.handle.net/1721.1/73114 https://orcid.org/0000-0001-8253-7714 https://orcid.org/0000-0001-5490-2323 |
| _version_ | 1811078964871954432 |
|---|---|
| author | Lis, Mieszko Shim, Keun Sup Cho, Myong Hyon Khan, Omer Devadas, Srinivas |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Lis, Mieszko Shim, Keun Sup Cho, Myong Hyon Khan, Omer Devadas, Srinivas |
| author_sort | Lis, Mieszko |
| collection | MIT |
| description | We introduce the concept of deadlock-free migration-based coherent shared memory to the NUCA family of architectures. Migration-based architectures move threads among cores to guarantee sequential semantics in large multicores. Using a execution migration (EM) architecture, we achieve performance comparable to directory-based architectures without using directories: avoiding automatic data replication significantly reduces cache miss rates, while a fast network-level thread migration scheme takes advantage of shared data locality to reduce remote cache accesses that limit traditional NUCA performance. EM area and energy consumption are very competitive, and, on the average, it outperforms a directory-based MOESI baseline by 1.3 x and a traditional S-NUCA design by 1.2 x. We argue that with EM scaling performance has much lower cost and design complexity than in directory based coherence and traditional NUCA architectures: by merely scaling network bandwidth from 256 to 512 bit flits, the performance of our architecture improves by an additional 13%, while the baselines show negligible improvement. |
| first_indexed | 2024-09-23T11:08:02Z |
| format | Article |
| id | mit-1721.1/73114 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T11:08:02Z |
| publishDate | 2012 |
| publisher | ACTA Press |
| record_format | dspace |
| spelling | mit-1721.1/731142022-10-01T01:28:14Z Directoryless shared memory coherence using execution migration Lis, Mieszko Shim, Keun Sup Cho, Myong Hyon Khan, Omer Devadas, Srinivas Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Lis, Mieszko Shim, Keun Sup Cho, Myong Hyon Khan, Omer Devadas, Srinivas We introduce the concept of deadlock-free migration-based coherent shared memory to the NUCA family of architectures. Migration-based architectures move threads among cores to guarantee sequential semantics in large multicores. Using a execution migration (EM) architecture, we achieve performance comparable to directory-based architectures without using directories: avoiding automatic data replication significantly reduces cache miss rates, while a fast network-level thread migration scheme takes advantage of shared data locality to reduce remote cache accesses that limit traditional NUCA performance. EM area and energy consumption are very competitive, and, on the average, it outperforms a directory-based MOESI baseline by 1.3 x and a traditional S-NUCA design by 1.2 x. We argue that with EM scaling performance has much lower cost and design complexity than in directory based coherence and traditional NUCA architectures: by merely scaling network bandwidth from 256 to 512 bit flits, the performance of our architecture improves by an additional 13%, while the baselines show negligible improvement. 2012-09-24T14:56:22Z 2012-09-24T14:56:22Z 2011-12 Article http://purl.org/eprint/type/ConferencePaper 1925-5543 http://hdl.handle.net/1721.1/73114 Lis, Mieszko et al. “Directoryless Shared Memory Coherence Using Execution Migration.” Parallel and Distributed Computing and Systems, December 2011. https://orcid.org/0000-0001-8253-7714 https://orcid.org/0000-0001-5490-2323 en_US http://dx.doi.org/10.2316/P.2011.757-081 Parallel and Distributed Computing and Systems Creative Commons Attribution-Noncommercial-Share Alike 3.0 http://creativecommons.org/licenses/by-nc-sa/3.0/ application/pdf ACTA Press MIT web domain |
| spellingShingle | Lis, Mieszko Shim, Keun Sup Cho, Myong Hyon Khan, Omer Devadas, Srinivas Directoryless shared memory coherence using execution migration |
| title | Directoryless shared memory coherence using execution migration |
| title_full | Directoryless shared memory coherence using execution migration |
| title_fullStr | Directoryless shared memory coherence using execution migration |
| title_full_unstemmed | Directoryless shared memory coherence using execution migration |
| title_short | Directoryless shared memory coherence using execution migration |
| title_sort | directoryless shared memory coherence using execution migration |
| url | http://hdl.handle.net/1721.1/73114 https://orcid.org/0000-0001-8253-7714 https://orcid.org/0000-0001-5490-2323 |
| work_keys_str_mv | AT lismieszko directorylesssharedmemorycoherenceusingexecutionmigration AT shimkeunsup directorylesssharedmemorycoherenceusingexecutionmigration AT chomyonghyon directorylesssharedmemorycoherenceusingexecutionmigration AT khanomer directorylesssharedmemorycoherenceusingexecutionmigration AT devadassrinivas directorylesssharedmemorycoherenceusingexecutionmigration |