Real-Time Data Bases

Such as several data processing systems and real-time systems often are coupled with a   database management function. Moreover, distributed databases would look to be a preferred approach in real-time systems because multitasking is commonplace and data are often process in sequence. If the database is distributed individual tasks can access their data faster and more reliably and with fewer bottlenecks than with a centralized database. Use of a distributed database for real-time applications divides input or output traffic and shortens queues of tasks waiting for access to a database. However, a failure of one database will rarely cause the failure of the entire system if redundancy is established in.

The performance efficiencies achieved by the use of a distributed database must be weighed against potential problems related with data partitioning and replication. While data redundancy improves response time through providing multiple information sources replication requirements for distributed files also produce logistical and overhead problems still all the files copies must be updated. In addition the use of distributed databases introduces the problem of concurrency control involves synchronizing the database so in which all copies have the identical information correct, free for access.

The conventional approach to concurrency control is   based on what are known as time stamps and locking.   At regular intervals the following tasks are initiated:

(1) the database is   locked so in which concurrency control is assured no Input or Output is permitted

(2) updating occurs as needed

(3) the database is unlocked

(4) files are validated to ensure in which all updates have been truly made

(5) the completed update is acknowledged. All locking tasks are monitored through a master clock example for time stamps. The delays involved in these process as well as the problems of avoiding deadlock inconsistent and updates militate against the widespread use of distributed databases.

Any  methods,  however,  have  been  established  to  speed  updating  and  to  solve  the  concurrency problem. 1 of these called the exclusive-writer protocol maintains the consistency of replicated files through allowing only a single exclusive writing task to update a file. It thus eliminates the high overhead of time stamp procedures or locking.