RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access

Chenchen Huang; Huiqi Hu; Xuecheng Qi; Xuan Zhou; Aoying Zhou

doi:10.1007/978-3-030-59410-7_22

RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access

Chenchen Huang, Huiqi Hu^*, Xuecheng Qi, Xuan Zhou, Aoying Zhou

^*Corresponding author for this work

School of Data Science and Engineering

East China Normal University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

Many key-value stores use RDMA to optimize the messaging and data transmission between application layer and storage layer, most of which only provide point-wise operations. Skiplist-based store can support both point operations and range queries, but its CPU-intensive access operations combined with the high-speed network will easily lead to the storage layer reaches CPU bottlenecks. In this paper, we present RS-store, a skiplist-based key-value store with RDMA, which can overcome the cpu handle of the storage layer by enabling two access modes: local access and remote access. In RS-store, we redesign a novel data structure R-skiplist to save the communication cost in remote access, and implement a latch-free concurrency control mechanism to ensure all the concurrency during two access modes. At last, our evaluation on a RDMA-capable cluster shows that the performance of RS-store over R-skiplist is 0.6 –1 higher than the existing skiplist, and it supports application layer’s high scalability.

Original language	English
Title of host publication	Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings
Editors	Yunmook Nah, Bin Cui, Sang-Won Lee, Jeffrey Xu Yu, Yang-Sae Moon, Steven Euijong Whang
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	314-323
Number of pages	10
ISBN (Print)	9783030594091
DOIs	https://doi.org/10.1007/978-3-030-59410-7_22
State	Published - 2020
Event	25th International Conference on Database Systems for Advanced Applications, DASFAA 2020 - Jeju, Korea, Republic of Duration: 24 Sep 2020 → 27 Sep 2020

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12112 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th International Conference on Database Systems for Advanced Applications, DASFAA 2020
Country/Territory	Korea, Republic of
City	Jeju
Period	24/09/20 → 27/09/20

Keywords

Key-value store
RDMA
Skiplist

Access to Document

10.1007/978-3-030-59410-7_22

Cite this

Huang, C., Hu, H., Qi, X., Zhou, X., & Zhou, A. (2020). RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access. In Y. Nah, B. Cui, S.-W. Lee, J. X. Yu, Y.-S. Moon, & S. E. Whang (Eds.), Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings (pp. 314-323). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12112 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-59410-7_22

Huang, Chenchen ; Hu, Huiqi ; Qi, Xuecheng et al. / RS-store : A SkipList-Based Key-Value Store with Remote Direct Memory Access. Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. editor / Yunmook Nah ; Bin Cui ; Sang-Won Lee ; Jeffrey Xu Yu ; Yang-Sae Moon ; Steven Euijong Whang. Springer Science and Business Media Deutschland GmbH, 2020. pp. 314-323 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{eea8986f3f5d4541a8fe66e9e41f83ae,

title = "RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access",

abstract = "Many key-value stores use RDMA to optimize the messaging and data transmission between application layer and storage layer, most of which only provide point-wise operations. Skiplist-based store can support both point operations and range queries, but its CPU-intensive access operations combined with the high-speed network will easily lead to the storage layer reaches CPU bottlenecks. In this paper, we present RS-store, a skiplist-based key-value store with RDMA, which can overcome the cpu handle of the storage layer by enabling two access modes: local access and remote access. In RS-store, we redesign a novel data structure R-skiplist to save the communication cost in remote access, and implement a latch-free concurrency control mechanism to ensure all the concurrency during two access modes. At last, our evaluation on a RDMA-capable cluster shows that the performance of RS-store over R-skiplist is 0.6 –1 higher than the existing skiplist, and it supports application layer{\textquoteright}s high scalability.",

keywords = "Key-value store, RDMA, Skiplist",

author = "Chenchen Huang and Huiqi Hu and Xuecheng Qi and Xuan Zhou and Aoying Zhou",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 25th International Conference on Database Systems for Advanced Applications, DASFAA 2020 ; Conference date: 24-09-2020 Through 27-09-2020",

year = "2020",

doi = "10.1007/978-3-030-59410-7\_22",

language = "英语",

isbn = "9783030594091",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "314--323",

editor = "Yunmook Nah and Bin Cui and Sang-Won Lee and Yu, \{Jeffrey Xu\} and Yang-Sae Moon and Whang, \{Steven Euijong\}",

booktitle = "Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings",

address = "德国",

}

Huang, C, Hu, H, Qi, X, Zhou, X & Zhou, A 2020, RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access. in Y Nah, B Cui, S-W Lee, JX Yu, Y-S Moon & SE Whang (eds), Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12112 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 314-323, 25th International Conference on Database Systems for Advanced Applications, DASFAA 2020, Jeju, Korea, Republic of, 24/09/20. https://doi.org/10.1007/978-3-030-59410-7_22

RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access. / Huang, Chenchen; Hu, Huiqi; Qi, Xuecheng et al.
Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. ed. / Yunmook Nah; Bin Cui; Sang-Won Lee; Jeffrey Xu Yu; Yang-Sae Moon; Steven Euijong Whang. Springer Science and Business Media Deutschland GmbH, 2020. p. 314-323 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12112 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - RS-store

T2 - 25th International Conference on Database Systems for Advanced Applications, DASFAA 2020

AU - Huang, Chenchen

AU - Hu, Huiqi

AU - Qi, Xuecheng

AU - Zhou, Xuan

AU - Zhou, Aoying

PY - 2020

Y1 - 2020

N2 - Many key-value stores use RDMA to optimize the messaging and data transmission between application layer and storage layer, most of which only provide point-wise operations. Skiplist-based store can support both point operations and range queries, but its CPU-intensive access operations combined with the high-speed network will easily lead to the storage layer reaches CPU bottlenecks. In this paper, we present RS-store, a skiplist-based key-value store with RDMA, which can overcome the cpu handle of the storage layer by enabling two access modes: local access and remote access. In RS-store, we redesign a novel data structure R-skiplist to save the communication cost in remote access, and implement a latch-free concurrency control mechanism to ensure all the concurrency during two access modes. At last, our evaluation on a RDMA-capable cluster shows that the performance of RS-store over R-skiplist is 0.6 –1 higher than the existing skiplist, and it supports application layer’s high scalability.

AB - Many key-value stores use RDMA to optimize the messaging and data transmission between application layer and storage layer, most of which only provide point-wise operations. Skiplist-based store can support both point operations and range queries, but its CPU-intensive access operations combined with the high-speed network will easily lead to the storage layer reaches CPU bottlenecks. In this paper, we present RS-store, a skiplist-based key-value store with RDMA, which can overcome the cpu handle of the storage layer by enabling two access modes: local access and remote access. In RS-store, we redesign a novel data structure R-skiplist to save the communication cost in remote access, and implement a latch-free concurrency control mechanism to ensure all the concurrency during two access modes. At last, our evaluation on a RDMA-capable cluster shows that the performance of RS-store over R-skiplist is 0.6 –1 higher than the existing skiplist, and it supports application layer’s high scalability.

KW - Key-value store

KW - RDMA

KW - Skiplist

UR - https://www.scopus.com/pages/publications/85092107249

U2 - 10.1007/978-3-030-59410-7_22

DO - 10.1007/978-3-030-59410-7_22

M3 - 会议稿件

AN - SCOPUS:85092107249

SN - 9783030594091

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 314

EP - 323

BT - Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings

A2 - Nah, Yunmook

A2 - Cui, Bin

A2 - Lee, Sang-Won

A2 - Yu, Jeffrey Xu

A2 - Moon, Yang-Sae

A2 - Whang, Steven Euijong

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 24 September 2020 through 27 September 2020

ER -

Huang C, Hu H, Qi X, Zhou X , Zhou A. RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access. In Nah Y, Cui B, Lee SW, Yu JX, Moon YS, Whang SE, editors, Database Systems for Advanced Applications - 25th International Conference, DASFAA 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 314-323. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-59410-7_22

RS-store: A SkipList-Based Key-Value Store with Remote Direct Memory Access

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this