Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory
New data storage technologies such as the recently introduced Intel® Optane™ DC Persistent Memory Module (PMM) offer exciting opportunities for optimizing the query processing performance of database workloads. In particular, the unique combination of low latency, byte-addressability, persistence, a...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Association for Computing Machinery (ACM)
2021
|
| Online Access: | https://hdl.handle.net/1721.1/130104 |
| _version_ | 1826204874865377280 |
|---|---|
| author | Shanbhag, Anil Atmanand Tatbul Bitim, Emine Nesime Cohen, David Madden, Samuel R |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Shanbhag, Anil Atmanand Tatbul Bitim, Emine Nesime Cohen, David Madden, Samuel R |
| author_sort | Shanbhag, Anil Atmanand |
| collection | MIT |
| description | New data storage technologies such as the recently introduced Intel® Optane™ DC Persistent Memory Module (PMM) offer exciting opportunities for optimizing the query processing performance of database workloads. In particular, the unique combination of low latency, byte-addressability, persistence, and large capacity make persistent memory (PMem) an attractive alternative along with DRAM and SSDs. Exploring the performance characteristics of this new medium is the first critical step in understanding how it will impact the design and performance of database systems. In this paper, we present one of the first experimental studies on characterizing Intel® Optane™ DC PMM's performance behavior in the context of analytical database workloads. First, we analyze basic access patterns common in such workloads, such as sequential, selective, and random reads as well as the complete Star Schema Benchmark, comparing standalone DRAM- and PMem-based implementations. Then we extend our analysis to join algorithms over larger datasets, which require using DRAM and PMem in a hybrid fashion while paying special attention to the read-write asymmetry of PMem. Our study reveals interesting performance tradeoffs that can help guide the design of next-generation OLAP systems in presence of persistent memory in the storage hierarchy. |
| first_indexed | 2024-09-23T13:02:27Z |
| format | Article |
| id | mit-1721.1/130104 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T13:02:27Z |
| publishDate | 2021 |
| publisher | Association for Computing Machinery (ACM) |
| record_format | dspace |
| spelling | mit-1721.1/1301042022-09-28T11:41:49Z Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory Shanbhag, Anil Atmanand Tatbul Bitim, Emine Nesime Cohen, David Madden, Samuel R Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory New data storage technologies such as the recently introduced Intel® Optane™ DC Persistent Memory Module (PMM) offer exciting opportunities for optimizing the query processing performance of database workloads. In particular, the unique combination of low latency, byte-addressability, persistence, and large capacity make persistent memory (PMem) an attractive alternative along with DRAM and SSDs. Exploring the performance characteristics of this new medium is the first critical step in understanding how it will impact the design and performance of database systems. In this paper, we present one of the first experimental studies on characterizing Intel® Optane™ DC PMM's performance behavior in the context of analytical database workloads. First, we analyze basic access patterns common in such workloads, such as sequential, selective, and random reads as well as the complete Star Schema Benchmark, comparing standalone DRAM- and PMem-based implementations. Then we extend our analysis to join algorithms over larger datasets, which require using DRAM and PMem in a hybrid fashion while paying special attention to the read-write asymmetry of PMem. Our study reveals interesting performance tradeoffs that can help guide the design of next-generation OLAP systems in presence of persistent memory in the storage hierarchy. 2021-03-08T21:38:09Z 2021-03-08T21:38:09Z 2020-06 Article http://purl.org/eprint/type/ConferencePaper 9781450380249 https://hdl.handle.net/1721.1/130104 Shanbhag, Anil et al. "Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory." DaMoN '20: Proceedings of the 16th International Workshop on Data Management on New Hardware, June 2020, Portland, Oregon, Association for Computing Machinery, June 2020. http://dx.doi.org/10.1145/3399666.3399933 DaMoN '20: Proceedings of the 16th International Workshop on Data Management on New Hardware Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Association for Computing Machinery (ACM) Prof. Madden via Phoebe Ayers |
| spellingShingle | Shanbhag, Anil Atmanand Tatbul Bitim, Emine Nesime Cohen, David Madden, Samuel R Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory |
| title | Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory |
| title_full | Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory |
| title_fullStr | Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory |
| title_full_unstemmed | Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory |
| title_short | Large-scale in-memory analytics on Intel ® Optane™ DC persistent memory |
| title_sort | large scale in memory analytics on intel r optane™ dc persistent memory |
| url | https://hdl.handle.net/1721.1/130104 |
| work_keys_str_mv | AT shanbhaganilatmanand largescaleinmemoryanalyticsoninteloptanedcpersistentmemory AT tatbulbitimeminenesime largescaleinmemoryanalyticsoninteloptanedcpersistentmemory AT cohendavid largescaleinmemoryanalyticsoninteloptanedcpersistentmemory AT maddensamuelr largescaleinmemoryanalyticsoninteloptanedcpersistentmemory |