On the necessity of explicit cross-layer data formats in near-data processing systems
- Massive data transfers in modern data-intensive systems resulting from low data-locality and data-to-code system design hurt their performance and scalability. Near-Data processing (NDP) and a shift to code-to-data designs may represent a viable solution as packaging combinations of storage and compute elements on the same device has become feasible. The shift towards NDP system architectures calls for revision of established principles. Abstractions such as data formats and layouts typically spread multiple layers in traditional DBMS, the way they are processed is encapsulated within these layers of abstraction. The NDP-style processing requires an explicit definition of cross-layer data formats and accessors to ensure in-situ executions optimally utilizing the properties of the underlying NDP storage and compute elements. In this paper, we make the case for such data format definitions and investigate the performance benefits under RocksDB and the COSMOS hardware platform.
Author of HS Reutlingen | Petrov, Ilia; Vinçon, Tobias; Knödler, Christian; Bernhardt, Arthur |
---|---|
URN: | urn:nbn:de:bsz:rt2-opus4-38067 |
DOI: | https://doi.org/10.1007/s10619-021-07328-z |
ISSN: | 0926-8782 |
eISSN: | 1573-7578 |
Erschienen in: | Distributed and parallel databases : an international journal |
Publisher: | Springer Nature |
Place of publication: | Cham |
Document Type: | Journal article |
Language: | English |
Publication year: | 2022 |
Tag: | FPGA; data format; data layout; near-data processing |
Volume: | 40 |
Issue: | Special issue on self-managing and hardware-optimized database systems 2020 |
Page Number: | 19 |
First Page: | 27 |
Last Page: | 45 |
DDC classes: | 004 Informatik |
Open access?: | Ja |
Licence (German): | Creative Commons - CC BY - Namensnennung 4.0 International |