Data updating method based on updating marks and related equipment

Abstract

The invention relates to the field of big data, and discloses a data updating method based on an updating mark and related equipment, and the method comprises the steps: presetting a data updating condition, and obtaining a data updating request; when the to-be-updated data corresponding to the data updating request meets the preset data updating condition, querying whether an updating mark is setfor the to-be-updated data corresponding to the data updating request or not in a cache; and if the cache queries that the to-be-updated data corresponding to the data updating request is not provided with the updating mark or the updating mark is overtime, setting the updating mark for the to-be-updated data in the cache, applying for exclusive locking for the to-be-updated data, and performingdata updating according to a result of applying for exclusive locking. According to the data updating method and device, the updating mark is set for the updated data, exclusive locking is applied forthe data with the updating mark, data updating is conducted, and system performance reduction caused by data deadlock can be avoided.

Description

technical field [0001] This application relates to the field of big data, in particular to a data update method and related equipment based on an update mark. Background technique [0002] In asynchronous scheduling in information systems, data update errors often occur. In a large-scale system, if data deviations lead to errors, it may cause serious problems and it is difficult to trace the root cause, and bring huge losses to users. Therefore, when data is updated, an exclusive lock is usually used to ensure the accuracy of the data in the information system. The exclusive lock means that when the data with the exclusive lock is updated, other processes cannot modify it. The data, until the data releases the exclusive lock, but because in the process of using the exclusive lock, once multiple objects apply for the exclusive lock, but if the exclusive lock is cross-occupied, it will cause The situation of exclusive lock deadlock, which leads to performance degradation; for...

AI Technical Summary

Problems solved by technology

[0002] In asynchronous scheduling in information systems, data update errors often occur. In a large-scale system, if data deviations lead to errors, it may cause serious problems and it is difficult to trace the root cause, and bring huge losses to users. Therefore, when data is updated, an exclusive lock is usually used to ensure the accuracy of the data in the information system. The exclusive lock means that when the data with the exclusive lock is updated, other processes cannot modify it. The data, until the data releases the exclusive lock, but because in the process of using the exclusive lock, once multiple objects apply for the exclusive lock, but if the exclusive lock is cross-occupied, it will cause The situation of exclusive lock deadlock, which leads to performance degradation; for example: suppose there are resources A and B, program X and program Y both update the two resources of A and B at the same time, so they must preempt the exclusive resources of A and B However, during the application process, program X seizes the exclusive lock of resource A, and program Y seizes the exclusive lock of resource B. At this time, program X will wait for the exclusive lock of resource B, and program Y will wait for the exclusive lock of resource A. Exclusive lock, obviously neither program X nor program Y can wait, which will lead to deadlock, so a new way is needed to apply for exclusive lock to solve the problem of deadlock

Images