Archiving Techniques for Temporal Databases

This paper describes archiving strategies for append-only temporal databases. We present a storage architecture where optical disks work in tandem with magnetic disks. Magnetic disks are used for storing current versions and recent past versions, whereas optical disks are dedicated for archiving older past versions. Similarly, temporal access structures are stored on both magnetic and optical disks. is limited. Another reason is that unlike current data, historical data is less likely to be referenced, and therefore a relatively slower medium such as optical disks is suitable for on-line archiving of historical data. Our migration techniques: (1) allow temporal data and access structures to span magnetic disks and optical disks; (2) minimize the overhead of the migration process by taking advantage of append-only nature of temporal databases; (3) gracefully handle object versions with very long time intervals so that the delay in the migration process is kept to minimum; and (4) ensure that no false magnetic or optical disk address lookup is performed during search operations by duplicating some closed versions on both magnetic and optical disks. Write-Once Read Many (WORM) optical disks provide a cost-effective, efficient solution for organizations requiring permanent storage for large amounts of temporal data. They offer the ability to store massive quantities of data in a relatively small amount of space because of lower cost of storage per byte. They have the advantage of not suffering from deterioration unlike magnetic media; a 100 year lifespan has been predicted for WORM disks. Although the access time of optical disks is higher than that of magnetic disks, WORM disks are still considered on-line random access devices, and they offer the best choice for archiving immutable data. To validate our claims for the efficiency of migration techniques, we analyze the performance of temporal access structures partitioned between magnetic and optical disks. We show that the migration process has a minimal effect on the search time. Our simulation identifies important parameters, and shows how they affect the performance of the temporal access structures. These include mean of version lifespan, block size, query time interval length, and total number of versions. Database systems take periodic backups to provide recovery from disk failures. These backups reflect some past state of the application. Whenever a failure occurs, the backups together with the log are used to restore the database to a consistent state. In temporal databases, past states of the application are kept to provide on-line access. Although this capability can be achieved by keeping full backups of the application, the performance of the resulting system will definitely be very poor. To overcome this deficiency, the issues of automatic archiving and searching of both object versions and temporal access structures must be successfully addressed, which is the topic of this paper.