The Journal of Information Science and Technology Association
(Johono Kagaku to Gijutsu)
Abstract

Masaru KITSUREGAWA
Institute of Industrial Science, the University of Tokyo
4-6-1 Komaba Meguro-Ku, Tokyo 153-8505, JAPAN

Abstract: In this article, the brief history of database technologies is given. Relational database gave a substantial revolution in data management. The core of database technologies such as transaction processing and query processing etc. are addressed. The recent technology trends such as probabilistic database, privacy, XML database, data mining, stream processing are introduced. The author tries to overlook the future, especially the IT system architecture of big data era.

Keywords：  database / data mining / stream processing / storage system / cloud computing / Big Data

Masakazu HATTORI, Hitoshi TANIGAWA, Yuji KONDO
Toshiba Solutions Corp. Platform Solutions Div.
1-1-15, Musashidai, Fuchu-shi, Tokyo 183-8532, JAPAN

Abstract: XML is one of the most widely-used formats which expresses various data universally for electronic-government, electronic commerce, office work data, newspaper article and so on. As XML applications spread out, the need to manage XML data in its native format increases in enterprise environments. It is an XML database to satisfy the needs.
In this report, we speak basic thought and operation method about an XML database with the flexibility inherent in XML. And, we argue about a merit of the XML database utilization through examples of the document management application development.

Keywords：  XML / database / XMLDB / document management / native XML database

Ikki OHMUKAI
National Institute of Informatics
2-1-2 Hitotsubashi, Chiyoda-ku, Tokyo 101-8430, JAPAN

Abstract: In order to process large amount of access at low cost, it is necessary to design the system considering important functions to be provided. In CiNii Articles, we have achieved the performance requirements by using search engine and simple DBMS. We also provide reliable information service by the bibliographic ID management system.

Keywords：  system design / search engine / DBMS / unique ID / disambiguation

Tomoaki SUGIYAMA^＊1
Fumi MORIUCHI^＊2
Rie TAKAHASHI^＊1
Keisuke MORIBE^＊1
＊1Shizuoka University Library
836, Ohya, Suruga-ku, Shizuoka city, Shizuoka Pref, 422-8529 JAPAN
^＊2 Shizuoka University Library (former affiliation)

Abstract: With the growth of Cloud Computing, more and more companies are offering different cloud services for libraries, and libraries are adopting cloud-based library management services. Shizuoka University has renovated its information infrastructure and chosen to use both private and public cloud services. At the same time, Shizuoka University Library has driven all systems and services, including the academic repository and the management system, into the cloud. Backup and restore options have been carefully evaluated during the transition period. Cloud Computing releases us from hosting local servers and equipment and constantly dealing with hardware failures. When electricity went out because of a huge typhoon, we have realized how important it is to organize disaster plans.

Keywords：  cloud computing / information infrastructure / library management system / academic repository system / data maintenance

Yoji KIYOTA
Littel Laboratory, NEXT Co., Ltd.
Shinagawa Front Bldg. 2-3-13 Konan, Minato-ku, Tokyo, 108-0075 JAPAN

Abstract: Relational Database Systems (RDBMS), which have been used as a core component for modern information services, have important features such as transaction and security. However, RDBMS have some issues for use of big data, i.e. terabytes or petabytes of data, such as log data of large-scale web information services. Distributed computing systems, such as Hadoop and distributed key-value stores, are spreading for use of big data. This article describes mechanisms of distributed computing systems, and proper use of RDBMS and distributed computing systems. And then, this article shows some consideration of methodologies for use of big data at organizations such as libraries.

Keywords：  big data / RDBMS / distributed computing / Hadoop / distributed key-value store