Systems and Methods for Managing the Synchronization of Replicated Version-Managed Databases

Abstract

A system and method for synchronizing a local replicated database with remote replicated databases. Generally, the system includes a local replicated database and a synchronization manager associated with the local replicated database. The synchronization manager may send changes made on the local replicated database to one or more remote replicated databases for reconstruction by the one or more remote replicated databases, and may receive changes made on a remote replicated database. In addition, the synchronization manager may reconstruct changes received from a remote replicated database on the local replicated database. Preferably, the synchronization manager may send, receive, and reconstruct changes independently from one another, i.e., may perform one or more of these activities autonomously and / or asynchronously.

Description

[0001] This application is a continuation of co-pending application Ser. No. 09 / 991,485, filed Nov. 13, 2001.FIELD OF THE INVENTION [0002] The present invention relates generally to replicated version-managed databases, and more particularly to systems and methods for managing the synchronization of replicated version-managed databases. BACKGROUND OF THE INVENTION [0003] A database transaction is a group of related updates against a database that form a single “unit of work.” There are basically two types of transactions: “short transactions” and “long transactions.” Short transactions typically take a few seconds or less to complete. A simple example is a bank account transfer, which may involve subtracting the appropriate amount from an entry in one account and adding the same amount to another account. Since this may take a relatively short period of time to complete, the records may be locked while the transaction is in progress to ensure that nobody else is able to change them ...

AI Technical Summary

Benefits of technology

[0020] Thus, each of the databases in a replicated database network may include a synchronization manager, thereby allowing each database to autonomously and / or asynchronously exchange changes and / or reconstruct changes independent of other databases in the system. Another advantage of the synchronization manager is that it may monitor activities of the local database with which it is associated and selectively perform one or more of its activities when it most efficient to do so. For example, the synchronization manager may send and / or receive changes when the interface of the local database is available, e.g., not being used for other tasks. In addition or alternatively, the synchronization manager may reconstruct changes, e.g., to synchronize the local database with other databases, when the local database has resources, e.g., processor or memory capacity, available or at times that substantially minimize interference with operation of the local database. Thus, the synchronization manager may operate substantially undetected by users of the local database, while still maintaining the local database substantially synchronized with other databases in the network.

Problems solved by technology

In contrast, a long transaction may take hours, weeks, or months.

Because these transactions may take a long period of time, locking records for an environment with multiple users who need to access the same data is inefficient and counterproductive.

Concurrent control becomes a critical issue.

In the case where the network 5 connecting the version managed databases 1 is unreliable or high-speed communications between database 1 sites is unavailable, single process synchronization may be inefficient.

Images