HiRDB adopts a Shared Nothing architecture able to maintain high performance even for increasing amounts of data, just by adding new servers.
increasing amounts of data handled as a result of business growth. This means that the
IT systems face processing performance demanded for IT systems is growing significantly. Since HiRDB enables processing performance to grow proportionally to the number of servers, it enables high performance to be maintained even when the amount of data increases. This is due to its "Shared Nothing" architecture, in which resources are never shared across multiple servers.
Since there is no competition over each server resource, it can be applied to even large-scale systems of more than 10 servers.
Systems can start on a small scale, and then be developed into large-scale systems, enabling operation to continue seamlessly to suit the scope of business.
The amount of data processed by business systems increases every year. This has resulted in a greater amount of time being taken up by batch processing to handle the data, as well as leading to reduced online operation time. HiRDB carries out batch processing quickly to extend the time available for online operation.
HiRDB can logically split up a table, and store the logical subsets on multiple servers across different locations-such as various business facilities-to process them in a distributed and parallel way. HiRDB can link with the grid batch system to make sets out of the data that has been stored on each of the respective servers and the batches that process that data.
This makes it possible to quickly process the batches on the different servers concurrently.
In addition, HiRDB has in-memory data processing capabilities that allows it to execute batch processing in memory. This further accelerates batch processing by eliminating the disk I/O processes that can create performance bottlenecks.
In this way, HiRDB can process even large amounts of data in a shorter time period, which means online business operation time can be extended.
See larger picture in a new window
XML is highly flexible, and widely used as an inter-operational interface and content exchange format. With XML, even when the data structure used for inter-operational interfaces or data structure for content exchange no longer needs to be changed, system changes can be kept to a minimum, to minimize impact on operations. HiRDB supports XML natively. The HiRDB XML format can be used to store XML data as-is in tables, without redefinition in the event that the XML data structure changes, taking maximum advantage of the flexibility of XML. HiRDB also supports most of the functionality for SQL/XML, an SQL standard that uses XQuery, a standard XML query language.
