Performance and resource modeling in highly-concurrent OLTP workloads
Database administrators of Online Transaction Processing (OLTP) systems constantly face difcult questions. For example, "What is the maximum throughput I can sustain with my current hardware?", "How much disk I/O will my system perform if the requests per second double?", or &quo...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
ACM Press
2021
|
| Online Access: | https://hdl.handle.net/1721.1/137854 |
| _version_ | 1826200186375897088 |
|---|---|
| author | Mozafari, Barzan Curino, Carlo Jindal, Alekh Madden, Samuel |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Mozafari, Barzan Curino, Carlo Jindal, Alekh Madden, Samuel |
| author_sort | Mozafari, Barzan |
| collection | MIT |
| description | Database administrators of Online Transaction Processing (OLTP) systems constantly face difcult questions. For example, "What is the maximum throughput I can sustain with my current hardware?", "How much disk I/O will my system perform if the requests per second double?", or "What will happen if the ratio of transactions in my system changes?". Resource prediction and performance analysis are both vital and difcult in this setting. Here the challenge is due to high degrees of concurrency, competition for resources, and complex interactions between transactions, all of which non-linearly impact performance. Although difcult, such analysis is a key component in enabling database administrators to understand which queries are eating up the resources, and how their system would scale under load. In this paper, we introduce our framework, called DBSeer, that addresses this problem by employing statistical models that provide resource and performance analysis and prediction for highly concurrent OLTP workloads. Our models are built on a small amount of training data from standard log information collected during normal system operation. Tese models are capable of accurately measuring several performance metrics, including resource consumption on a per-transaction-type basis, resource bottlenecks, and throughput at diferent load levels. We have validated these models on MySQL/Linux with numerous experiments on standard benchmarks (TPC-C) and real workloads (Wikipedia), observing high accuracy (within a few percent error) when predicting all of the above metrics. Copyright © 2013 ACM. |
| first_indexed | 2024-09-23T11:32:29Z |
| format | Article |
| id | mit-1721.1/137854 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T11:32:29Z |
| publishDate | 2021 |
| publisher | ACM Press |
| record_format | dspace |
| spelling | mit-1721.1/1378542023-04-14T16:06:49Z Performance and resource modeling in highly-concurrent OLTP workloads Mozafari, Barzan Curino, Carlo Jindal, Alekh Madden, Samuel Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Database administrators of Online Transaction Processing (OLTP) systems constantly face difcult questions. For example, "What is the maximum throughput I can sustain with my current hardware?", "How much disk I/O will my system perform if the requests per second double?", or "What will happen if the ratio of transactions in my system changes?". Resource prediction and performance analysis are both vital and difcult in this setting. Here the challenge is due to high degrees of concurrency, competition for resources, and complex interactions between transactions, all of which non-linearly impact performance. Although difcult, such analysis is a key component in enabling database administrators to understand which queries are eating up the resources, and how their system would scale under load. In this paper, we introduce our framework, called DBSeer, that addresses this problem by employing statistical models that provide resource and performance analysis and prediction for highly concurrent OLTP workloads. Our models are built on a small amount of training data from standard log information collected during normal system operation. Tese models are capable of accurately measuring several performance metrics, including resource consumption on a per-transaction-type basis, resource bottlenecks, and throughput at diferent load levels. We have validated these models on MySQL/Linux with numerous experiments on standard benchmarks (TPC-C) and real workloads (Wikipedia), observing high accuracy (within a few percent error) when predicting all of the above metrics. Copyright © 2013 ACM. 2021-11-09T13:21:13Z 2021-11-09T13:21:13Z 2013 2019-06-18T13:18:33Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137854 Mozafari, Barzan, Curino, Carlo, Jindal, Alekh and Madden, Samuel. 2013. "Performance and resource modeling in highly-concurrent OLTP workloads." en 10.1145/2463676.2467800 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf ACM Press other univ website |
| spellingShingle | Mozafari, Barzan Curino, Carlo Jindal, Alekh Madden, Samuel Performance and resource modeling in highly-concurrent OLTP workloads |
| title | Performance and resource modeling in highly-concurrent OLTP workloads |
| title_full | Performance and resource modeling in highly-concurrent OLTP workloads |
| title_fullStr | Performance and resource modeling in highly-concurrent OLTP workloads |
| title_full_unstemmed | Performance and resource modeling in highly-concurrent OLTP workloads |
| title_short | Performance and resource modeling in highly-concurrent OLTP workloads |
| title_sort | performance and resource modeling in highly concurrent oltp workloads |
| url | https://hdl.handle.net/1721.1/137854 |
| work_keys_str_mv | AT mozafaribarzan performanceandresourcemodelinginhighlyconcurrentoltpworkloads AT curinocarlo performanceandresourcemodelinginhighlyconcurrentoltpworkloads AT jindalalekh performanceandresourcemodelinginhighlyconcurrentoltpworkloads AT maddensamuel performanceandresourcemodelinginhighlyconcurrentoltpworkloads |