Faghih, Faezeh ; István, Zsolt ; Dinu, Florin
Hrsg.: Bordawekar, Rajesh ; Lahiri, Tirthankar (2022)
A Short Study of Recent Smart Storage Solutions for OLAP: Lessons and Opportunities.
48th International Conference on Very Large Databases. Sydney, Australia (05.-09.09.2022)
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
As the usage of data-intensive applications is increasing, moving data between the storage and host is becoming an important bottleneck. Smart storage has emerged as a solution to mitigate the data-movement bottleneck, resulting in lower latency, better energy efficiency, and better overall throughput. The last decade has seen a proliferation of proposals for architectures for smart storage. In this short study, we look at a representative sample of these works and distill common lessons and techniques from them: first, high selectivity in itself is not a guarantee for better performance, second, it is important to match the chosen in-storage data layout with that of the processing element and application and third, leveraging heterogeneous processing elements inside smart storage benefits efficiency and flexibility. Our main goal in compiling and sharing the list of lessons is to avoid reinventing the wheel when designing future smart storage solutions. At the same time, we also identify aspects that will be necessary for building future smart storage that can cater to varied workloads but are not well covered in related work today.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2022 |
| Herausgeber: | Bordawekar, Rajesh ; Lahiri, Tirthankar |
| Autor(en): | Faghih, Faezeh ; István, Zsolt ; Dinu, Florin |
| Art des Eintrags: | Bibliographie |
| Titel: | A Short Study of Recent Smart Storage Solutions for OLAP: Lessons and Opportunities |
| Sprache: | Englisch |
| Publikationsjahr: | 5 September 2022 |
| Buchtitel: | International Workshop on Accelerating Analytics and Data Management Systems Using Modern Processor and Storage Architectures, ADMS@VLDB 2022 |
| Veranstaltungstitel: | 48th International Conference on Very Large Databases |
| Veranstaltungsort: | Sydney, Australia |
| Veranstaltungsdatum: | 05.-09.09.2022 |
| Zugehörige Links: | |
| Kurzbeschreibung (Abstract): | As the usage of data-intensive applications is increasing, moving data between the storage and host is becoming an important bottleneck. Smart storage has emerged as a solution to mitigate the data-movement bottleneck, resulting in lower latency, better energy efficiency, and better overall throughput. The last decade has seen a proliferation of proposals for architectures for smart storage. In this short study, we look at a representative sample of these works and distill common lessons and techniques from them: first, high selectivity in itself is not a guarantee for better performance, second, it is important to match the chosen in-storage data layout with that of the processing element and application and third, leveraging heterogeneous processing elements inside smart storage benefits efficiency and flexibility. Our main goal in compiling and sharing the list of lessons is to avoid reinventing the wheel when designing future smart storage solutions. At the same time, we also identify aspects that will be necessary for building future smart storage that can cater to varied workloads but are not well covered in related work today. |
| Freie Schlagworte: | systems_funding_96000590, systems_huawei_ndpda |
| Zusätzliche Informationen: | 13th International Workshop on Accelerating Analytics and Data Management Systems Using Modern Processor and Storage Architectures |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Distributed and Networked Systems |
| Hinterlegungsdatum: | 07 Nov 2023 11:00 |
| Letzte Änderung: | 07 Nov 2023 11:00 |
| PPN: | |
| Zugehörige Links: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum