Home > Research > Publications & Outputs > Gyro: A Modular Scale-Out Layer for Single-Serv...

Research

Associated organisational units

Text available via DOI:

Keywords

View graph of relations

Gyro: A Modular Scale-Out Layer for Single-Server DBMSs

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review

Published

Standard

Gyro: A Modular Scale-Out Layer for Single-Server DBMSs. / Suri, Neeraj.
2019 38th Symposium on Reliable Distributed Systems (SRDS). Vol. 2019 Lyon, France, 2020. p. 71-82.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review

Harvard

Suri, N 2020, Gyro: A Modular Scale-Out Layer for Single-Server DBMSs. in 2019 38th Symposium on Reliable Distributed Systems (SRDS). vol. 2019, Lyon, France, pp. 71-82. https://doi.org/10.1109/SRDS47363.2019.00018

APA

Suri, N. (2020). Gyro: A Modular Scale-Out Layer for Single-Server DBMSs. In 2019 38th Symposium on Reliable Distributed Systems (SRDS) (Vol. 2019, pp. 71-82). https://doi.org/10.1109/SRDS47363.2019.00018

Vancouver

Suri N. Gyro: A Modular Scale-Out Layer for Single-Server DBMSs. In 2019 38th Symposium on Reliable Distributed Systems (SRDS). Vol. 2019. Lyon, France. 2020. p. 71-82 doi: 10.1109/SRDS47363.2019.00018

Author

Suri, Neeraj. / Gyro: A Modular Scale-Out Layer for Single-Server DBMSs. 2019 38th Symposium on Reliable Distributed Systems (SRDS). Vol. 2019 Lyon, France, 2020. pp. 71-82

Bibtex

@inproceedings{1d4fc9caa66b4f14b883e5756cc0f547,
title = "Gyro: A Modular Scale-Out Layer for Single-Server DBMSs",
abstract = "Scaling out database management systems (DBMSs) requires distributed coordination, which can easily become a bottleneck. Recent work on speeding up distributed transactions has addressed this problem by proposing scale-out techniques that are deeply integrated with the concurrency control mechanism of the DBMS. This paper explores the design of modular coordination layers, which encapsulate all scale-out logic and can be applied to scale out any unmodified single-server DBMS. It proposes Gyro, a modular coordination layer that runs on top of a collection of single-server DBMS instances and interacts with them only through their client interface. Gyro distributes the load by ensuring that as many requests as possible are executed by only one DBMS instance. Our experiments show that modular distributed coordination is practically viable and that it can be much faster than traditional distributed transaction protocols using two-phase commit.",
keywords = "Concurrency control, Protocols, Servers, Instruments, Electronic commerce, Standards, Database systems",
author = "Neeraj Suri",
year = "2020",
month = mar,
day = "30",
doi = "10.1109/SRDS47363.2019.00018",
language = "English",
volume = "2019",
pages = "71--82",
booktitle = "2019 38th Symposium on Reliable Distributed Systems (SRDS)",

}

RIS

TY - GEN

T1 - Gyro: A Modular Scale-Out Layer for Single-Server DBMSs

AU - Suri, Neeraj

PY - 2020/3/30

Y1 - 2020/3/30

N2 - Scaling out database management systems (DBMSs) requires distributed coordination, which can easily become a bottleneck. Recent work on speeding up distributed transactions has addressed this problem by proposing scale-out techniques that are deeply integrated with the concurrency control mechanism of the DBMS. This paper explores the design of modular coordination layers, which encapsulate all scale-out logic and can be applied to scale out any unmodified single-server DBMS. It proposes Gyro, a modular coordination layer that runs on top of a collection of single-server DBMS instances and interacts with them only through their client interface. Gyro distributes the load by ensuring that as many requests as possible are executed by only one DBMS instance. Our experiments show that modular distributed coordination is practically viable and that it can be much faster than traditional distributed transaction protocols using two-phase commit.

AB - Scaling out database management systems (DBMSs) requires distributed coordination, which can easily become a bottleneck. Recent work on speeding up distributed transactions has addressed this problem by proposing scale-out techniques that are deeply integrated with the concurrency control mechanism of the DBMS. This paper explores the design of modular coordination layers, which encapsulate all scale-out logic and can be applied to scale out any unmodified single-server DBMS. It proposes Gyro, a modular coordination layer that runs on top of a collection of single-server DBMS instances and interacts with them only through their client interface. Gyro distributes the load by ensuring that as many requests as possible are executed by only one DBMS instance. Our experiments show that modular distributed coordination is practically viable and that it can be much faster than traditional distributed transaction protocols using two-phase commit.

KW - Concurrency control

KW - Protocols

KW - Servers

KW - Instruments

KW - Electronic commerce

KW - Standards

KW - Database systems

U2 - 10.1109/SRDS47363.2019.00018

DO - 10.1109/SRDS47363.2019.00018

M3 - Conference contribution/Paper

VL - 2019

SP - 71

EP - 82

BT - 2019 38th Symposium on Reliable Distributed Systems (SRDS)

CY - Lyon, France

ER -