TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines
The rise in computing hardware choices is driving a reevaluation of operating systems. The traditional role of an operating system controlling the execution of its own hardware is evolving toward a model whereby the controlling processor is distinct from the compute engines that are performing most...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2020
|
| Online Access: | https://hdl.handle.net/1721.1/126114 |
| _version_ | 1826200664477270016 |
|---|---|
| author | Kepner, Jeremy Brightwell, Ron Edelman, Alan Gadepally, Vijay N. Hayden, Jananthan Jones, Michael Madden, Samuel R. Michaleas, Peter W. Okhravi, Hamed Pedretti, Kevin Reuther, Albert I. Sterling, Thomas Stonebraker, Michael |
| author2 | Lincoln Laboratory |
| author_facet | Lincoln Laboratory Kepner, Jeremy Brightwell, Ron Edelman, Alan Gadepally, Vijay N. Hayden, Jananthan Jones, Michael Madden, Samuel R. Michaleas, Peter W. Okhravi, Hamed Pedretti, Kevin Reuther, Albert I. Sterling, Thomas Stonebraker, Michael |
| author_sort | Kepner, Jeremy |
| collection | MIT |
| description | The rise in computing hardware choices is driving a reevaluation of operating systems. The traditional role of an operating system controlling the execution of its own hardware is evolving toward a model whereby the controlling processor is distinct from the compute engines that are performing most of the computations. In this context, an operating system can be viewed as software that brokers and tracks the resources of the compute engines and is akin to a database management system. To explore the idea of using a database in an operating system role, this work defines key operating system functions in terms of rigorous mathematical semantics (associative array algebra) that are directly translatable into database operations. These operations possess a number of mathematical properties that are ideal for parallel operating systems by guaranteeing correctness over a wide range of parallel operations. The resulting operating system equations provide a mathematical specification for a Tabular Operating System Architecture (TabulaROSA) that can be implemented on any platform. Simulations of forking in TabularROSA are performed using an associative array implementation and compared to Linux on a 32,000+ core supercomputer. Using over 262,000 forkers managing over 68,000,000,000 processes, the simulations show that TabulaROSA has the potential to perform operating system functions on a massively parallel scale. The TabulaROSA simulations show 20x higher performance as compared to Linux while managing 2000x more processes in fully searchable tables. |
| first_indexed | 2024-09-23T11:39:56Z |
| format | Article |
| id | mit-1721.1/126114 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T11:39:56Z |
| publishDate | 2020 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1261142022-09-27T21:06:27Z TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines Kepner, Jeremy Brightwell, Ron Edelman, Alan Gadepally, Vijay N. Hayden, Jananthan Jones, Michael Madden, Samuel R. Michaleas, Peter W. Okhravi, Hamed Pedretti, Kevin Reuther, Albert I. Sterling, Thomas Stonebraker, Michael Lincoln Laboratory Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory The rise in computing hardware choices is driving a reevaluation of operating systems. The traditional role of an operating system controlling the execution of its own hardware is evolving toward a model whereby the controlling processor is distinct from the compute engines that are performing most of the computations. In this context, an operating system can be viewed as software that brokers and tracks the resources of the compute engines and is akin to a database management system. To explore the idea of using a database in an operating system role, this work defines key operating system functions in terms of rigorous mathematical semantics (associative array algebra) that are directly translatable into database operations. These operations possess a number of mathematical properties that are ideal for parallel operating systems by guaranteeing correctness over a wide range of parallel operations. The resulting operating system equations provide a mathematical specification for a Tabular Operating System Architecture (TabulaROSA) that can be implemented on any platform. Simulations of forking in TabularROSA are performed using an associative array implementation and compared to Linux on a 32,000+ core supercomputer. Using over 262,000 forkers managing over 68,000,000,000 processes, the simulations show that TabulaROSA has the potential to perform operating system functions on a massively parallel scale. The TabulaROSA simulations show 20x higher performance as compared to Linux while managing 2000x more processes in fully searchable tables. United States. Department of Defense. Assistant Secretary of Defense for Research & Engineering (Air Force Contract No. FA8721-05-C-0002) United States. Department of Defense. Assistant Secretary of Defense for Research & Engineering (Air Force Contract No. FA8702-15-D-0001) 2020-07-09T14:24:56Z 2020-07-09T14:24:56Z 2018-11 2018-09 2019-06-18T17:22:29Z Article http://purl.org/eprint/type/ConferencePaper 2377-6943 INSPEC Accession Number: 18290388 https://hdl.handle.net/1721.1/126114 Kepner, Jeremy, Ron Brightwell, Alan Edelman et al. in Proceedings of the 2018 IEEE High Performance Extreme Computing Conference (HPEC), 25-27 Sept. 2018, Waltham, MA, USA, © 2018 IEEE. en https://dx.doi.org/10.1109/HPEC.2018.8547577 High Performance Extreme Computing Conference (HPEC), 2018 IEEE Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) arXiv |
| spellingShingle | Kepner, Jeremy Brightwell, Ron Edelman, Alan Gadepally, Vijay N. Hayden, Jananthan Jones, Michael Madden, Samuel R. Michaleas, Peter W. Okhravi, Hamed Pedretti, Kevin Reuther, Albert I. Sterling, Thomas Stonebraker, Michael TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines |
| title | TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines |
| title_full | TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines |
| title_fullStr | TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines |
| title_full_unstemmed | TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines |
| title_short | TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines |
| title_sort | tabularosa tabular operating system architecture for massively parallel heterogeneous compute engines |
| url | https://hdl.handle.net/1721.1/126114 |
| work_keys_str_mv | AT kepnerjeremy tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT brightwellron tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT edelmanalan tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT gadepallyvijayn tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT haydenjananthan tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT jonesmichael tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT maddensamuelr tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT michaleaspeterw tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT okhravihamed tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT pedrettikevin tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT reutheralberti tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT sterlingthomas tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines AT stonebrakermichael tabularosatabularoperatingsystemarchitectureformassivelyparallelheterogeneouscomputeengines |