Method for processing database affair, client, and server

Abstract

The invention provides a method for processing a database affair, a client and a server. The method includes: allowing the server to receive a database access request for requesting target data and sent by the client; sending a first version of the target data to the client, wherein the first version is a version of the target data which is currently stored in the server, and the server only savesthe last submitted updated version of the target data; receiving an affair submitting request sent by the client who executes a first affair on the basis of the first version, wherein the affair submitting request includes a log for recording data operation of the first affair; modifying the version of the target data stored in the server currently according to the data operation of the first affair if the data operation of the first affair and the version of the target data stored in the server currently do not have data conflicts according to the log, and acquiring a submitted updated version of the target data. The method for processing the database affair can achieve an unlocked affair mechanism, and can lower the complexity of storage management of the database.

Description

technical field [0001] The present invention relates to the technical field of databases, and more specifically, to a method for processing database transactions, a client and a server. Background technique [0002] When multiple users access the database at the same time, multiple transactions will operate on the same data concurrently. If such concurrent operations are not controlled, the problem of accessing incorrect data in the database may occur, destroying the relationship between the transaction and the database. consistency. [0003] Blocking is a very important technique to achieve concurrency control. The so-called blocking means that transaction T sends a request to the system to lock it before operating on a certain data. After locking, transaction T has certain control over the data. Before transaction T releases its lock, other transactions are locked. The data cannot be manipulated. The size of a blocked object is called the blocked granularity. Blocking ...

AI Technical Summary

Problems solved by technology

[0007] As can be seen from the above, compared with the traditional lock mechanism, the existing MVCC mechanism avoids reading blocking writing and writing blocking reading, but it still cannot avoid writing blocking writing.

In addition, in the existing MVCC technology, due to the need to store multiple versions of the same data, the storage management of the database is more complicated, and it is also necessary to regularly clean up the no longer needed versions to reclaim space, which also increases additional overhead

Images