Graphite: A Distributed Parallel Simulator for Multicores
This paper introduces the open-source Graphite distributed parallel multicore simulator infrastructure. Graphite is designed from the ground up for exploration of future multicore processors containing dozens, hundreds, or even thousands of cores. It provides high performance for fast design space e...
| Main Authors: | , , , , , , , |
|---|---|
| Outros Autores: | |
| Publicado em: |
2009
|
| Acesso em linha: | http://hdl.handle.net/1721.1/49809 |
| _version_ | 1826192125253910528 |
|---|---|
| author | Beckmann, Nathan Eastep, Jonathan Gruenwald, Charles, III Kurian, George Kasture, Harshad Miller, Jason E. Celio, Christopher Agarwal, Anant |
| author2 | Anant Agarwal |
| author_facet | Anant Agarwal Beckmann, Nathan Eastep, Jonathan Gruenwald, Charles, III Kurian, George Kasture, Harshad Miller, Jason E. Celio, Christopher Agarwal, Anant |
| author_sort | Beckmann, Nathan |
| collection | MIT |
| description | This paper introduces the open-source Graphite distributed parallel multicore simulator infrastructure. Graphite is designed from the ground up for exploration of future multicore processors containing dozens, hundreds, or even thousands of cores. It provides high performance for fast design space exploration and software development for future processors. Several techniques are used to achieve this performance including: direct execution, multi-machine distribution, analytical modeling, and lax synchronization. Graphite is capable of accelerating simulations by leveraging several machines. It can distribute simulation of an off-the-shelf threaded application across a cluster of commodity Linux machines with no modification to the source code. It does this by providing a single, shared address space and consistent single-process image across machines. Graphite is designed to be a simulation framework, allowing different component models to be easily replaced to either model different architectures or tradeoff accuracy for performance. We evaluate Graphite from a number of perspectives and demonstrate that it can simulate target architectures containing over 1000 cores on ten 8-core servers. Performance scales well as more machines are added with near linear speedup in many cases. Simulation slowdown is as low as 41x versus native execution for some applications. The Graphite infrastructure and existing models will be released as open-source software to allow the community to simulate their own architectures and extend and improve the framework. |
| first_indexed | 2024-09-23T09:06:39Z |
| id | mit-1721.1/49809 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T09:06:39Z |
| publishDate | 2009 |
| record_format | dspace |
| spelling | mit-1721.1/498092019-04-10T20:48:51Z Graphite: A Distributed Parallel Simulator for Multicores Beckmann, Nathan Eastep, Jonathan Gruenwald, Charles, III Kurian, George Kasture, Harshad Miller, Jason E. Celio, Christopher Agarwal, Anant Anant Agarwal Computer Architecture This paper introduces the open-source Graphite distributed parallel multicore simulator infrastructure. Graphite is designed from the ground up for exploration of future multicore processors containing dozens, hundreds, or even thousands of cores. It provides high performance for fast design space exploration and software development for future processors. Several techniques are used to achieve this performance including: direct execution, multi-machine distribution, analytical modeling, and lax synchronization. Graphite is capable of accelerating simulations by leveraging several machines. It can distribute simulation of an off-the-shelf threaded application across a cluster of commodity Linux machines with no modification to the source code. It does this by providing a single, shared address space and consistent single-process image across machines. Graphite is designed to be a simulation framework, allowing different component models to be easily replaced to either model different architectures or tradeoff accuracy for performance. We evaluate Graphite from a number of perspectives and demonstrate that it can simulate target architectures containing over 1000 cores on ten 8-core servers. Performance scales well as more machines are added with near linear speedup in many cases. Simulation slowdown is as low as 41x versus native execution for some applications. The Graphite infrastructure and existing models will be released as open-source software to allow the community to simulate their own architectures and extend and improve the framework. 2009-11-10T18:15:03Z 2009-11-10T18:15:03Z 2009-11-09 http://hdl.handle.net/1721.1/49809 MIT-CSAIL-TR-2009-056 17 p. application/pdf |
| spellingShingle | Beckmann, Nathan Eastep, Jonathan Gruenwald, Charles, III Kurian, George Kasture, Harshad Miller, Jason E. Celio, Christopher Agarwal, Anant Graphite: A Distributed Parallel Simulator for Multicores |
| title | Graphite: A Distributed Parallel Simulator for Multicores |
| title_full | Graphite: A Distributed Parallel Simulator for Multicores |
| title_fullStr | Graphite: A Distributed Parallel Simulator for Multicores |
| title_full_unstemmed | Graphite: A Distributed Parallel Simulator for Multicores |
| title_short | Graphite: A Distributed Parallel Simulator for Multicores |
| title_sort | graphite a distributed parallel simulator for multicores |
| url | http://hdl.handle.net/1721.1/49809 |
| work_keys_str_mv | AT beckmannnathan graphiteadistributedparallelsimulatorformulticores AT eastepjonathan graphiteadistributedparallelsimulatorformulticores AT gruenwaldcharlesiii graphiteadistributedparallelsimulatorformulticores AT kuriangeorge graphiteadistributedparallelsimulatorformulticores AT kastureharshad graphiteadistributedparallelsimulatorformulticores AT millerjasone graphiteadistributedparallelsimulatorformulticores AT celiochristopher graphiteadistributedparallelsimulatorformulticores AT agarwalanant graphiteadistributedparallelsimulatorformulticores |