HiRDB Datareplicator Version 8 Description, User's Guide and Operator's Guide
![[Contents]](FIGURE/CONTENT.GIF)
![[Glossary]](FIGURE/GLOSS.GIF)
![[Index]](FIGURE/INDEX.GIF)
![[Back]](FIGURE/FRONT.GIF)
![[Next]](FIGURE/AFTER.GIF)
- abstract data type
- A data type for large and complicated data, such as multimedia data or three-dimensional data. Abstract data types comply with the SQL3 standard with object-oriented extensions. These data types enable users to define unique data with a complicated structure as well as procedures for its manipulation.
- Datareplicator supports HiRDB's SGMLTEXT, FREEWORD, and XML abstract data types for data linkage. To use the SGMLTEXT and FREEWORD types, you must have HiRDB Text Search Plug-in or HiRDB XML Extension. HiRDB XML Extension is required to use the XML type,.
- ADT (Abstract Data Type)
- See abstract data type.
- advanced queue
- A message queuing function provided by Oracle. For details, see the Oracle documentation.
- application
- Collective name for HiRDB application processing. A program created as an application is referred to as the application program (or user application program (UAP)).
- association
- Logical communication route used to establish communication using the OSI protocol.
- back-end server
- A server component of a HiRDB/Parallel Server. A back-end server provides functions for accessing and locking databases, and functions for receiving and executing sort/merge tasks from other back-end servers.
- BES (Back-end Server)
- See back-end server.
- BLOB (Binary Large Object)
- Acronym for Binary Large Object, which refers to large data, such as documents, images, and audio data.
- client/server
- Term indicating the relationship of program-to-program communications. A client issues an application processing request, and a server accepts the request and executes the application. Client/server indicates a relative relationship between such programs. The term client/server can be applied to programs as well as to hardware (such as workstations and personal computers).
- cluster key
- A table column that is specified as the key for storing rows in ascending or descending order of the values in the specified column.
- data definition language
- A language for defining the organization and contents of a database.
- data dictionary
- A dictionary that contains a database's design specifications, such as the database table structures, column definitions, and index definitions. A single database is divided for storage purposes among multiple servers (in the case of HiRDB/Parallel Server) and managed collectively.
- data dictionary LOB RDAREA
- An RDAREA for storing stored procedures. The two types of data dictionary LOB RDAREAs are a type for storing the definition sources of stored procedures and a type for storing their objects.
- data manipulation language
- A language for specifying how a database will be manipulated by an application program.
- data warehouse
- A mechanism for managing a key system's data in such a manner that the data can be used easily by end users in a data system.
- A key application's databases are designed to achieve efficient, high-speed data update processing. For the data application's databases, on the other hand, managing data in an easy-to-handle format according to the end user's purposes is more important than processing efficiency and speed. A data warehouse provides data applications with an effective data utilization mechanism. When a data warehouse is constructed, the data accumulated by key applications can be used efficiently by data applications, such as for end users' decision making support and application analysis. Additionally, if a database is created to accumulate updated data in chronological order (a time-ordered database), trends and progressive changes in information can also be analyzed.
- Datareplicator and HiRDB Dataextractor's replication facility enable a HiRDB server to be used as a database in a data warehouse.
- deadlock
- Deadlock occurs when multiple transactions are competing for the same resources and are waiting for each other to release their resources.
- dictionary server
- A server component of a HiRDB/Parallel Server. A dictionary server is used to manage data dictionaries.
- DS (Dictionary Server)
- See dictionary server.
- embedded UAP
- A UAP in which SQL statements are written directly into a source program that is written in a high-level language (C or COBOL).
- FES (Front-end Server)
- See front-end server.
- FIX attribute table
- A table whose rows are fixed in length.
- flexible hash partitioning
- Method for partitioning a table that uses a hash function to effect uniform distribution of the table's column values among RDAREAs. The column used as the basis for partitioning a table's rows is called the partitioning key.
- FREEWORD type
- See abstract data type.
- front-end server
- A server component of a HiRDB/Parallel Server. A front-end server provides functions for establishing connections with clients, distributing processing to back-end servers, and collecting the processing results.
- global buffer
- Buffer used for input/output of data between a HiRDB and the HiRDB files.
- global buffer pool
- A set of global buffer sectors of the same size. A global buffer is used for input/output of data stored in RDAREAs on the disk. A global buffer pool is allocated in shared memory. Global buffers are always allocated for RDAREAs and indexes.
- host
- A single computer (machine) connect to a network on which HiRDB is run. In the case of a HiRDB/Parallel Server, servers are distributed to multiple hosts. In the case of a multi-HiRDB, one host consists of multiple HiRDBs.
- index
- An index created on the basis of specific columns in a table in order to improve data retrieval efficiency. An index is a structure assigned to one or more columns to provide a key for searching the table; an index consists of the key and key values. A key is a column name that indicates the contents of the column, while key values are the data values set in the column.
- The two types of indexes are single-column indexes and multicolumn indexes. A single-column index is based on indexing only one column of a table. A multicolumn index is based on indexing multiple columns of a table as a single index.
- interface area
- An area used to pass information between HiRDB and a UAP.
- JP1
- Collective name for a group of products equipped with various facilities such as batch job operation, automatic system operation, forms output control, and file backup. JP1 is used to implement automatic system operation and energy conservation. The JP1 products include JP1/Automatic Operation Monitor, which automates system operation, and JP1/System Event Service, which manages events issued by programs.
- key range partitioning
- A method of partitioning a table. Key range partitioning divides the table into groups of rows on the basis of ranges of values in a specified column of the table. The column that is the basis for row partitioning is called the partitioning key.
- linkage recovery backup file
- File created for each HiRDB file by using the backup command (pdfbkup) to back up HiRDB files containing unload log files from the HiRDB file system area. Linkage recovery backup files are used by the data linkage recovery facility.
- LOB data
- Large, variable-length data, such as documents, images, and audio LOB data is stored in user LOB RDAREAs. Data loading and database reorganization can be executed for LOB RDAREAs separately from the user RDAREAs containing the LOB column structure base table.
- MGR (System Manager)
- See system manager.
- multi-HiRDB
- A mode in which multiple HiRDBs run on a single host. Each HiRDB is operated separately. A separate HiRDB administrator must be registered for each HiRDB.
- node
- A communication control unit and the system connected to it that constitute a network component in a distributed system. A node refers to a host or a HiRDB system.
- null value
- A special value indicating that no value has been set.
- partitioning key index
- An index whose primary component column is the one for which storage conditions are specified for a partitioned table. It includes the following indexes:
- A single-column index created on the basis of the partitioning key
- A multicolumn index created on the basis of multiple columns beginning with the partitioning key
- An index that is not a partitioning key index is called a non-partitioning key index.
- plug-in
- A package product that extends HiRDB's data management capabilities. HiRDB's abstract data types allow a wide variety of data to be retrieved. HiRDB provides the following plug-ins:
- HiRDB Text Search Plug-in
- HiRDB Image Search Plug-in
- HiRDB Spatial Search Plug-in
- port number
- A number assigned to a service that is available in a network.
- process
- The unit of program execution in a work area that is created when application software, such as a UAP or HiRDB, is executed under the operating system. Virtual space and time-shared CPU resources are allocated to each process.
- RD-node
- A DBMS identified by a location in a network that is connected with the distributed database facility. A node name used to specify a DBMS at the target server in the distributed network is referred to as the RD-node name. The RD-node name is equivalent to the resource name under the OSI-RDA protocol.
- regular file
- A simple UNIX dynamic file. Because it can be created and deleted with UNIX commands, a regular file is suitable for creating a test database.
- repetition column
- A column composed of multiple elements in a relational database table. Because a repetition column can store multiple elements in a single cell, what were conventionally separate tables can be combined into a single table.
- Datareplicator data linkage handles a repetition column as a single column. Data linkage is not supported for each element of a repetition column.
- replication facility
- A facility that enables data to be imported from one database into a database in another system by linking databases between a mainframe and HiRDB, or between two separate HiRDBs. HiRDB's replication facilities include the data linkage facility (HiRDB Datareplicator) and the data extraction and import service facility (HiRDB Dataextractor).
- row partitioning
- Division of a table, index, or LOB column into multiple partitions for storage in multiple user RDAREAs or user LOB RDAREAs. When a table is row partitioned, its indexes can also be partitioned on the same basis as the table partitions. If a table contains a LOB column, the LOB column can be divided and stored in multiple user LOB RDAREAs on the same basis as the table partitions.
- server
- In the HiRDB manuals, a server refers to a component unit constituting a HiRDB system.
- SGMLTEXT type
- See abstract data type.
- statistics log file
- A file for storing statistical information (statistics log) output by HiRDB.
- status file
- A file used to store system status information that is needed in order to restart HiRDB.
- superuser
- The UNIX user with the highest privileges. A superuser can access all files in the UNIX file system. The superuser name is root.
- synchronization point
- A punctuation point of transaction processing. A commit refers to the synchronization point processing that indicates that transaction processing terminated normally, while a rollback refers to the synchronization point processing that is executed in order to ignore transaction processing because it did not execute successfully.
- synchronization point dump file
- A file containing HiRDB management information needed to restart HiRDB or restore a database. The system log is separated by synchronization points at specified intervals, and HiRDB management information is saved at each synchronization point. In the event of a failure, the system is restored by using the most recent synchronization point dump and the system log information that has been collected since that synchronization point dump occurred. This method reduces system recovery time because there is no need to read the collected system log from the beginning.
- system files
- A collective name for the following types of files:
- System log files
- Synchronization point dump files
- Status files
- system log file
- A file used to store historical information about database update processing.
- system manager
- A server component of a HiRDB/Parallel Server. The system manager's principal role is to control the execution of commands and utilities.
- system RDAREA
- A collective name for the following types of RDAREAs:
- Master directory RDAREA
- Data directory RDAREA
- Data dictionary RDAREAs
- TCP/IP
- A protocol developed by ARPANET, a project of Defense Advanced Research Projects Agency (DARPA). The TCP/IP protocol is used mainly with LANs.
- transaction
- To maintain data conformity, processing such as reading data from a file, and then writing the modified data (update processing) is handled as a unit and is not divided into segments. This unit of processing is called a transaction. The result of transaction processing is always either valid or invalid.
- trigger
- An operation that causes a predefined SQL statement to be executed automatically when an operation (updating, insertion, or deletion) is performed on a table. Triggers can be used to update tables automatically when a related table is updated.
- UAP
- An application created as a program; also referred to as an application program.
- unload log file
- A file created by unloading a system log file.
- unload statistics log file
- A file created by unloading the contents of a statistics log file.
- update information
- Information indicating the details of updating operations performed on the source database.
- user LOB RDAREA
- An RDAREA for storing BLOB data (large data, such as documents, images, and sounds).
- view table
- A virtual table defined by selecting rows and columns from an actual table (which is called the base table).
- volume
- The unit of allocation for storing segments.
- work table file
- A file used to temporarily store information that is required in order to execute SQL statements.
- XML type
- See abstract data type.
All rights reserved. Copyright (C) 2007, 2013, Hitachi, Ltd.