Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination

Hebatalla Eldakiky; David Hung Chang Du

doi:10.1109/MASCOTS53633.2021.9614283

Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination

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

Abstract

The power and flexibility of software-defined networks lead to a programmable network infrastructure in which in-network computation can help accelerating the performance of applications. This can be achieved by offloading some computational tasks to the network. In this paper, we propose TurboKV, an efficient distributed key-value store architecture that utilizes programmable switches as: 1) partition management nodes to store the key-value store partitions and replicas information; and 2) monitoring stations to measure and balance the load among storage nodes. We also propose a key-based routing protocol to route the queries of clients based on the requested keys to targeted storage nodes. Our experimental results of an initial prototype show that our proposed architecture improves the throughput and reduces the latency of distributed key-value stores when compared to the existing architectures.

Original language	English (US)
Title of host publication	Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021
Publisher	IEEE Computer Society
ISBN (Electronic)	9781665458382
DOIs	https://doi.org/10.1109/MASCOTS53633.2021.9614283
State	Published - 2021
Externally published	Yes
Event	29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021 - Houston, United States Duration: Nov 3 2021 → Nov 5 2021

Publication series

Name	Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS
ISSN (Print)	1526-7539

Conference

Conference	29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021
Country/Territory	United States
City	Houston
Period	11/3/21 → 11/5/21

Bibliographical note

Funding Information:
X. ACKNOWLEDGMENT This work was supported by NSF I/UCRC Center Research in Intelligent Storage and the NSF awards 1439662 and 1812537.

Publisher Copyright:
© 2021 IEEE.

Access

10.1109/MASCOTS53633.2021.9614283

OpenUrl availability

Full text

Cite this

Eldakiky, H., & Du, D. H. C. (2021). Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination. In Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021 (Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS). IEEE Computer Society. https://doi.org/10.1109/MASCOTS53633.2021.9614283

Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination. / Eldakiky, Hebatalla; Du, David Hung Chang.
Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021. IEEE Computer Society, 2021. (Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS).

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

Eldakiky, H & Du, DHC 2021, Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination. in Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021. Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS, IEEE Computer Society, 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021, Houston, United States, 11/3/21. https://doi.org/10.1109/MASCOTS53633.2021.9614283

Eldakiky H, Du DHC. Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination. In Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021. IEEE Computer Society. 2021. (Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS). doi: 10.1109/MASCOTS53633.2021.9614283

Eldakiky, Hebatalla ; Du, David Hung Chang. / Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination. Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021. IEEE Computer Society, 2021. (Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS).

@inproceedings{a0467028752f40a9b699a33714e62c17,

title = "Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination",

abstract = "The power and flexibility of software-defined networks lead to a programmable network infrastructure in which in-network computation can help accelerating the performance of applications. This can be achieved by offloading some computational tasks to the network. In this paper, we propose TurboKV, an efficient distributed key-value store architecture that utilizes programmable switches as: 1) partition management nodes to store the key-value store partitions and replicas information; and 2) monitoring stations to measure and balance the load among storage nodes. We also propose a key-based routing protocol to route the queries of clients based on the requested keys to targeted storage nodes. Our experimental results of an initial prototype show that our proposed architecture improves the throughput and reduces the latency of distributed key-value stores when compared to the existing architectures.",

author = "Hebatalla Eldakiky and Du, {David Hung Chang}",

note = "Funding Information: X. ACKNOWLEDGMENT This work was supported by NSF I/UCRC Center Research in Intelligent Storage and the NSF awards 1439662 and 1812537. Publisher Copyright: {\textcopyright} 2021 IEEE.; 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021 ; Conference date: 03-11-2021 Through 05-11-2021",

year = "2021",

doi = "10.1109/MASCOTS53633.2021.9614283",

language = "English (US)",

series = "Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS",

publisher = "IEEE Computer Society",

booktitle = "Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021",

}

TY - GEN

T1 - Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination

AU - Eldakiky, Hebatalla

AU - Du, David Hung Chang

PY - 2021

Y1 - 2021

N2 - The power and flexibility of software-defined networks lead to a programmable network infrastructure in which in-network computation can help accelerating the performance of applications. This can be achieved by offloading some computational tasks to the network. In this paper, we propose TurboKV, an efficient distributed key-value store architecture that utilizes programmable switches as: 1) partition management nodes to store the key-value store partitions and replicas information; and 2) monitoring stations to measure and balance the load among storage nodes. We also propose a key-based routing protocol to route the queries of clients based on the requested keys to targeted storage nodes. Our experimental results of an initial prototype show that our proposed architecture improves the throughput and reduces the latency of distributed key-value stores when compared to the existing architectures.

AB - The power and flexibility of software-defined networks lead to a programmable network infrastructure in which in-network computation can help accelerating the performance of applications. This can be achieved by offloading some computational tasks to the network. In this paper, we propose TurboKV, an efficient distributed key-value store architecture that utilizes programmable switches as: 1) partition management nodes to store the key-value store partitions and replicas information; and 2) monitoring stations to measure and balance the load among storage nodes. We also propose a key-based routing protocol to route the queries of clients based on the requested keys to targeted storage nodes. Our experimental results of an initial prototype show that our proposed architecture improves the throughput and reduces the latency of distributed key-value stores when compared to the existing architectures.

UR - http://www.scopus.com/inward/record.url?scp=85123223476&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85123223476&partnerID=8YFLogxK

U2 - 10.1109/MASCOTS53633.2021.9614283

DO - 10.1109/MASCOTS53633.2021.9614283

M3 - Conference contribution

AN - SCOPUS:85123223476

T3 - Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS

BT - Proceedings - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021

PB - IEEE Computer Society

T2 - 29th International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2021

Y2 - 3 November 2021 through 5 November 2021

ER -

Scaling Up The Performance of Distributed Key-Value Stores With In-Switch Coordination

Abstract

Publication series

Conference

Bibliographical note

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this