Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation

Minato Omori, Hiroaki Nishi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Recently, data traffic on the Internet has increased due to the rapid growth of various Internet-based services. The convergence of user requests means that servers are overloaded. To solve this problem, service providers generally install multiple servers and distribute requests using a load balancer. The existing load balancing algorithms do not estimate the size of the load of unknown requests. However, the requested contents are heterogeneous and complex, so the size of the load is dependent on the servers and the contents of the requests. In this study, we propose a load balancing algorithm that distributes the requests based on estimates of the processing time, which avoids mismatches between the characteristics of servers and the request contents. The processing time for requests is estimated based on the requested contents by online machine learning, and a strategy to cover the latency of machine learning is proposed and partially conducted. To test the algorithm, we built a model of multiple database servers and performed an experiment using real log data for database requests. The simulation results showed that the proposed algorithm reduced the average processing time for requests by 94.5% compared with round robin and by 28.3% compared with least connections.

Original languageEnglish
Title of host publicationProceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages278-283
Number of pages6
ISBN (Electronic)9781538648292
DOIs
Publication statusPublished - 2018 Sep 24
Event16th IEEE International Conference on Industrial Informatics, INDIN 2018 - Porto, Portugal
Duration: 2018 Jul 182018 Jul 20

Other

Other16th IEEE International Conference on Industrial Informatics, INDIN 2018
CountryPortugal
CityPorto
Period18/7/1818/7/20

Fingerprint

Servers
Processing
Resource allocation
Learning systems
Internet
Data base
Experiments
World Wide Web
Machine learning
Load balancing

Keywords

  • Database server
  • Load balancing
  • Online machine learning
  • Processing time estimation

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Information Systems and Management
  • Industrial and Manufacturing Engineering

Cite this

Omori, M., & Nishi, H. (2018). Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation. In Proceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018 (pp. 278-283). [8471931] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INDIN.2018.8471931

Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation. / Omori, Minato; Nishi, Hiroaki.

Proceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 278-283 8471931.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Omori, M & Nishi, H 2018, Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation. in Proceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018., 8471931, Institute of Electrical and Electronics Engineers Inc., pp. 278-283, 16th IEEE International Conference on Industrial Informatics, INDIN 2018, Porto, Portugal, 18/7/18. https://doi.org/10.1109/INDIN.2018.8471931
Omori M, Nishi H. Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation. In Proceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 278-283. 8471931 https://doi.org/10.1109/INDIN.2018.8471931
Omori, Minato ; Nishi, Hiroaki. / Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation. Proceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 278-283
@inproceedings{6d8c14d03f024d04bad812bbedf1b57e,
title = "Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation",
abstract = "Recently, data traffic on the Internet has increased due to the rapid growth of various Internet-based services. The convergence of user requests means that servers are overloaded. To solve this problem, service providers generally install multiple servers and distribute requests using a load balancer. The existing load balancing algorithms do not estimate the size of the load of unknown requests. However, the requested contents are heterogeneous and complex, so the size of the load is dependent on the servers and the contents of the requests. In this study, we propose a load balancing algorithm that distributes the requests based on estimates of the processing time, which avoids mismatches between the characteristics of servers and the request contents. The processing time for requests is estimated based on the requested contents by online machine learning, and a strategy to cover the latency of machine learning is proposed and partially conducted. To test the algorithm, we built a model of multiple database servers and performed an experiment using real log data for database requests. The simulation results showed that the proposed algorithm reduced the average processing time for requests by 94.5{\%} compared with round robin and by 28.3{\%} compared with least connections.",
keywords = "Database server, Load balancing, Online machine learning, Processing time estimation",
author = "Minato Omori and Hiroaki Nishi",
year = "2018",
month = "9",
day = "24",
doi = "10.1109/INDIN.2018.8471931",
language = "English",
pages = "278--283",
booktitle = "Proceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Request Distribution for Heterogeneous Database Server Clusters with Processing Time Estimation

AU - Omori, Minato

AU - Nishi, Hiroaki

PY - 2018/9/24

Y1 - 2018/9/24

N2 - Recently, data traffic on the Internet has increased due to the rapid growth of various Internet-based services. The convergence of user requests means that servers are overloaded. To solve this problem, service providers generally install multiple servers and distribute requests using a load balancer. The existing load balancing algorithms do not estimate the size of the load of unknown requests. However, the requested contents are heterogeneous and complex, so the size of the load is dependent on the servers and the contents of the requests. In this study, we propose a load balancing algorithm that distributes the requests based on estimates of the processing time, which avoids mismatches between the characteristics of servers and the request contents. The processing time for requests is estimated based on the requested contents by online machine learning, and a strategy to cover the latency of machine learning is proposed and partially conducted. To test the algorithm, we built a model of multiple database servers and performed an experiment using real log data for database requests. The simulation results showed that the proposed algorithm reduced the average processing time for requests by 94.5% compared with round robin and by 28.3% compared with least connections.

AB - Recently, data traffic on the Internet has increased due to the rapid growth of various Internet-based services. The convergence of user requests means that servers are overloaded. To solve this problem, service providers generally install multiple servers and distribute requests using a load balancer. The existing load balancing algorithms do not estimate the size of the load of unknown requests. However, the requested contents are heterogeneous and complex, so the size of the load is dependent on the servers and the contents of the requests. In this study, we propose a load balancing algorithm that distributes the requests based on estimates of the processing time, which avoids mismatches between the characteristics of servers and the request contents. The processing time for requests is estimated based on the requested contents by online machine learning, and a strategy to cover the latency of machine learning is proposed and partially conducted. To test the algorithm, we built a model of multiple database servers and performed an experiment using real log data for database requests. The simulation results showed that the proposed algorithm reduced the average processing time for requests by 94.5% compared with round robin and by 28.3% compared with least connections.

KW - Database server

KW - Load balancing

KW - Online machine learning

KW - Processing time estimation

UR - http://www.scopus.com/inward/record.url?scp=85055505123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055505123&partnerID=8YFLogxK

U2 - 10.1109/INDIN.2018.8471931

DO - 10.1109/INDIN.2018.8471931

M3 - Conference contribution

AN - SCOPUS:85055505123

SP - 278

EP - 283

BT - Proceedings - IEEE 16th International Conference on Industrial Informatics, INDIN 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -