TU Darmstadt / ULB / TUbiblio

On the necessity of explicit cross-layer data formats in near-data processing systems

Weber, Lukas ; Vinçon, Tobias ; Knödler, Christian ; Solis-Vasquez, Leonardo ; Bernhardt, Arthur ; Petrov, Ilia ; Koch, Andreas (2024)
On the necessity of explicit cross-layer data formats in near-data processing systems.
In: Distributed and Parallel Databases, 2022, 40 (1)
doi: 10.26083/tuprints-00023519
Artikel, Zweitveröffentlichung, Verlagsversion

WarnungEs ist eine neuere Version dieses Eintrags verfügbar.

Kurzbeschreibung (Abstract)

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.

Typ des Eintrags: Artikel
Erschienen: 2024
Autor(en): Weber, Lukas ; Vinçon, Tobias ; Knödler, Christian ; Solis-Vasquez, Leonardo ; Bernhardt, Arthur ; Petrov, Ilia ; Koch, Andreas
Art des Eintrags: Zweitveröffentlichung
Titel: On the necessity of explicit cross-layer data formats in near-data processing systems
Sprache: Englisch
Publikationsjahr: 10 Dezember 2024
Ort: Darmstadt
Publikationsdatum der Erstveröffentlichung: März 2022
Ort der Erstveröffentlichung: New York
Verlag: Springer
Titel der Zeitschrift, Zeitung oder Schriftenreihe: Distributed and Parallel Databases
Jahrgang/Volume einer Zeitschrift: 40
(Heft-)Nummer: 1
DOI: 10.26083/tuprints-00023519
URL / URN: https://tuprints.ulb.tu-darmstadt.de/23519
Zugehörige Links:
Herkunft: Zweitveröffentlichung DeepGreen
Kurzbeschreibung (Abstract):

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.

Freie Schlagworte: Near-data processing, Data format, Data layout, FPGA
Status: Verlagsversion
URN: urn:nbn:de:tuda-tuprints-235192
Zusätzliche Informationen:

Part of a collection: Special Issue on Self-Managing and Hardware-Optimized Database Systems 2020

Sachgruppe der Dewey Dezimalklassifikatin (DDC): 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik
Fachbereich(e)/-gebiet(e): 20 Fachbereich Informatik
20 Fachbereich Informatik > Eingebettete Systeme und ihre Anwendungen
Hinterlegungsdatum: 10 Dez 2024 12:51
Letzte Änderung: 16 Jan 2025 12:13
PPN:
Export:
Suche nach Titel in: TUfind oder in Google

Verfügbare Versionen dieses Eintrags

Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen