Tebis: Index Shipping for Efficient Replication in LSM Key-Value Stores

Vardoulakis, Michalis; Saloustros, Giorgos; González Férez, Pilar; Bilas, Angelos

Publication:
Tebis: Index Shipping for Efficient Replication in LSM Key-Value Stores

dc.contributor.author	Vardoulakis, Michalis
dc.contributor.author	Saloustros, Giorgos
dc.contributor.author	González Férez, Pilar
dc.contributor.author	Bilas, Angelos
dc.contributor.department	Ingeniería y Tecnología de Computadores
dc.date.accessioned	2024-01-25T16:44:40Z
dc.date.available	2024-01-25T16:44:40Z
dc.date.created	2021-10-09
dc.date.issued	2022-04-05
dc.description.abstract	Key-value (KV) stores based on LSM tree have become a foundational layer in the storage stack of datacenters and cloud services. Current approaches for achieving reliability and availability favor reducing network traffic and send to replicas only new KV pairs. As a result, they perform costly compactions to reorganize data in both the primary and backup nodes, which increases device I/O traffic and CPU overhead, and eventually hurts overall system performance. In this paper we describe Tebis, an efficient LSM-based KV store that reduces I/O amplification and CPU overhead for maintaining the replica index. We use a primary-backup replication scheme that performs compactions only on the primary nodes and sends pre-built indexes to backup nodes, avoiding all compactions in backup nodes. Our approach includes an efficient mechanism to deal with pointer translation across nodes in the pre-built region index. Our results show that Tebis reduces pressure on backup nodes compared to performing full compactions: Throughput is increased by 1.1 − 1.48×, CPU efficiency is increased by 1.06 − 1.54×, and I/O amplification is reduced by 1.13 − 1.81×, without increasing server to server network traffic excessively (by up to 1.09 − 1.82×).	es
dc.format	application/pdf	es
dc.format.extent	14	es
dc.identifier.doi	https://doi.org/10.1145/3492321.3519572
dc.identifier.uri	http://hdl.handle.net/10201/137804
dc.language	eng	es
dc.publisher	Association for Computing Machinery (ACM)	es
dc.relation	- Ámbito del proyecto: Nacional - Agencia financiadora: Ministerio de Ciencia e Innovación - Convocatoria: Jan. 2019 - Dec. 2021 - Nombre del proyecto: Técnicas Innovadoras en Computación Especializada y de Altas Prestaciones - Código o número del acuerdo de subvención: RTI2018-098156-B-C53	es
dc.relation.ispartof	Seventeenth European Conference on Computer Systems (EuroSys ’22), April 5–8, 2022, RENNES, France. ACM, New York, NY, USA	es
dc.relation.publisherversion	https://web.archive.org/web/20220329144121id_/https://dl.acm.org/doi/pdf/10.1145/3492321.3519572	es
dc.rights	info:eu-repo/semantics/openAccess	es
dc.rights	Attribution-NonCommercial-NoDerivatives 4.0 Internacional	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/4.0/	*
dc.subject	Key-value stores	es
dc.subject	B-trees	es
dc.subject	Flash memory	es
dc.subject	Network design principles	es
dc.subject.other	CDU::0 - Generalidades.::00 - Ciencia y conocimiento. Investigación. Cultura. Humanidades.::004 - Ciencia y tecnología de los ordenadores. Informática.	es
dc.title	Tebis: Index Shipping for Efficient Replication in LSM Key-Value Stores	es
dc.type	info:eu-repo/semantics/conferenceObject	es
dspace.entity.type	Publication	es