Scalable Database Server, HiRDB Version 8 Description
HiRDB is used in a network environment consisting of client/server systems. A server system in which a database is installed is called a HiRDB server; a client system in which UAPs are developed and executed is called a HiRDB client. HiRDB servers and HiRDB clients are referred to as HiRDB systems. Figure 1-1 shows the configuration of a HiRDB system.
Figure 1-1 Configuration of a HiRDB system
The two types of HiRDB servers are the HiRDB/Single Server and the HiRDB/Parallel Server. You must select one of these two types, as appropriate to your system mode or the types of operations you will be performing.
A HiRDB server runs on one of the following platforms:
This manual describes the Windows version of HiRDB (which includes Windows 2000, Windows XP Professional, and Windows Server 2003).
You would use a HiRDB/Single Server for a database system that consists of a single server machine. Because of its stable processing performance and simple operations as compared with a HiRDB Parallel Server, a HiRDB/Single Server is well-suited for small and medium-sized databases.
A HiRDB/Parallel Server makes it possible to link multiple server machines into a single database system so that a table can be split up and stored on multiple server machines. Because it permits different servers to perform retrieval tasks in parallel, a HiRDB Parallel Server provides improved performance. In addition, its capability to add or update large amounts of data or back up a database concurrently means that a HiRDB Parallel Server is able to maintain high performance even when the database becomes very large.
The Shared Nothing method, which fully exploits the hardware's capabilities, enables you to maintain stable performance by increasing the number of servers when the volume of data to be handled increases. In addition, a HiRDB Parallel Server can achieve well-balanced parallel processing in a manner that prevents concentration of processing on a specific server by delegating sorting, joining, and similar large-overhead tasks to other servers that have smaller workloads.
The two types of HiRDB clients are the HiRDB/Developer's Kit and HiRDB/Run Time.
A HiRDB/client runs on one of the following platforms:
HiRDB/Developer's Kit is a program for developing (pre-processing, compiling, linking) and executing UAPs. As the development language for UAPs, you can use C, C++, COBOL85, OOCOBOL, COBOL 2002, or JavaTM (SQLJ).
Because a HiRDB server includes the HiRDB/Developer's Kit functions, a HiRDB/Developer's Kit is not necessary for developing or executing a UAP on a HiRDB server; a HiRDB/Developer's Kit is necessary, however, to develop or execute a UAP on a client.
HiRDB/Run Time is a run-time program that is used exclusively for executing previously created UAPs. HiRDB/Run Time cannot be used to develop (preprocess, compile, link) a UAP; it can only be used to execute a UAP.
At a HiRDB client, you can develop and execute UAPs that access XDM/RD E2 databases. Such UAPs can be used to access XDM/RD E2 databases from HiRDB clients. The function that performs this type of access is called the XDM/RD E2 connection facility. Figure 1-2 provides an overview of the XDM/RD E2 connection facility.
Figure 1-2 Overview of the XDM/RD E2 connection facility
All Rights Reserved. Copyright (C) 2007, Hitachi, Ltd.