Improving Dynamic Scaling Performance of Cassandra

Saneyasu YAMAGUCHI; Yuki MORIMITSU

doi:10.1587/transinf.2016DAP0009

Improving Dynamic Scaling Performance of Cassandra

Saneyasu YAMAGUCHI, Yuki MORIMITSU

Full Text Views

0

Share
Cite this

Summary :

Load size for a service on the Internet changes remarkably every hour. Thus, it is expected for service system scales to change dynamically according to load size. KVS (key-value store) is a scalable DBMS (database management system) widely used in largescale Internet services. In this paper, we focus on Cassandra, a popular open-source KVS implementation, and discuss methods for improving dynamic scaling performance. First, we evaluate node joining time, which is the time to complete adding a node to a running KVS system, and show that its bottleneck process is disk I/O. Second, we analyze disk accesses in the nodes and indicate that some heavily accessed files cause a large number of disk accesses. Third, we propose two methods for improving elasticity, which means decreasing node adding and removing time, of Cassandra. One method reduces disk accesses significantly by keeping the heavily accessed file in the page cache. The other method optimizes I/O scheduler behavior. Lastly, we evaluate elasticity of our methods. Our experimental results demonstrate that the methods can improve the scaling-up and scaling-down performance of Cassandra.

Publication: IEICE TRANSACTIONS on Information Vol.E100-D No.4 pp.682-692

Publication Date: 2017/04/01

Publicized: 2017/01/17

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2016DAP0009

Type of Manuscript: Special Section PAPER (Special Section on Data Engineering and Information Management)

Category

Authors

Saneyasu YAMAGUCHI
Kogakuin University
Yuki MORIMITSU
Kogakuin University

Keyword

KVS, key-value store, Cassandra, page cache

Cite this

Copy

Saneyasu YAMAGUCHI, Yuki MORIMITSU, "Improving Dynamic Scaling Performance of Cassandra" in IEICE TRANSACTIONS on Information, vol. E100-D, no. 4, pp. 682-692, April 2017, doi: 10.1587/transinf.2016DAP0009.
Abstract: Load size for a service on the Internet changes remarkably every hour. Thus, it is expected for service system scales to change dynamically according to load size. KVS (key-value store) is a scalable DBMS (database management system) widely used in largescale Internet services. In this paper, we focus on Cassandra, a popular open-source KVS implementation, and discuss methods for improving dynamic scaling performance. First, we evaluate node joining time, which is the time to complete adding a node to a running KVS system, and show that its bottleneck process is disk I/O. Second, we analyze disk accesses in the nodes and indicate that some heavily accessed files cause a large number of disk accesses. Third, we propose two methods for improving elasticity, which means decreasing node adding and removing time, of Cassandra. One method reduces disk accesses significantly by keeping the heavily accessed file in the page cache. The other method optimizes I/O scheduler behavior. Lastly, we evaluate elasticity of our methods. Our experimental results demonstrate that the methods can improve the scaling-up and scaling-down performance of Cassandra.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.2016DAP0009/_p

Copy

@ARTICLE{e100-d_4_682,
author={Saneyasu YAMAGUCHI, Yuki MORIMITSU, },
journal={IEICE TRANSACTIONS on Information},
title={Improving Dynamic Scaling Performance of Cassandra},
year={2017},
volume={E100-D},
number={4},
pages={682-692},
abstract={Load size for a service on the Internet changes remarkably every hour. Thus, it is expected for service system scales to change dynamically according to load size. KVS (key-value store) is a scalable DBMS (database management system) widely used in largescale Internet services. In this paper, we focus on Cassandra, a popular open-source KVS implementation, and discuss methods for improving dynamic scaling performance. First, we evaluate node joining time, which is the time to complete adding a node to a running KVS system, and show that its bottleneck process is disk I/O. Second, we analyze disk accesses in the nodes and indicate that some heavily accessed files cause a large number of disk accesses. Third, we propose two methods for improving elasticity, which means decreasing node adding and removing time, of Cassandra. One method reduces disk accesses significantly by keeping the heavily accessed file in the page cache. The other method optimizes I/O scheduler behavior. Lastly, we evaluate elasticity of our methods. Our experimental results demonstrate that the methods can improve the scaling-up and scaling-down performance of Cassandra.},
keywords={},
doi={10.1587/transinf.2016DAP0009},
ISSN={1745-1361},
month={April},}

Copy

TY - JOUR
TI - Improving Dynamic Scaling Performance of Cassandra
T2 - IEICE TRANSACTIONS on Information
SP - 682
EP - 692
AU - Saneyasu YAMAGUCHI
AU - Yuki MORIMITSU
PY - 2017
DO - 10.1587/transinf.2016DAP0009
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2017
AB - Load size for a service on the Internet changes remarkably every hour. Thus, it is expected for service system scales to change dynamically according to load size. KVS (key-value store) is a scalable DBMS (database management system) widely used in largescale Internet services. In this paper, we focus on Cassandra, a popular open-source KVS implementation, and discuss methods for improving dynamic scaling performance. First, we evaluate node joining time, which is the time to complete adding a node to a running KVS system, and show that its bottleneck process is disk I/O. Second, we analyze disk accesses in the nodes and indicate that some heavily accessed files cause a large number of disk accesses. Third, we propose two methods for improving elasticity, which means decreasing node adding and removing time, of Cassandra. One method reduces disk accesses significantly by keeping the heavily accessed file in the page cache. The other method optimizes I/O scheduler behavior. Lastly, we evaluate elasticity of our methods. Our experimental results demonstrate that the methods can improve the scaling-up and scaling-down performance of Cassandra.
ER -