Storage system write cache data issuing method and related components

Abstract

The invention discloses a method for issuing write cache data of a storage system. The method comprises the steps of determining a concurrent processing threshold for scheduling a transaction to a write cache; when write cache data issuing is started, judging whether the to-be-processed data volume of the write cache reaches a concurrent processing threshold value or not according to the issued data volume and the write cache completion data volume; if so, pausing the issuing of the data to the write cache; and if not, continuing to issue the data of the write cache. According to the method, an issuing strategy of data from the transaction to the write cache in a controllable mode is provided; the concurrent threshold from transaction scheduling to write cache IO is introduced; and the transaction IO issuing thread is scheduled only when the concurrent number is smaller than the threshold value, so that the data processing pressure of the write cache is reduced; the situation that theIO of an upper-layer host is overtime is avoided; and the user experience is improved. The invention further provides a device and equipment for issuing the write cache data of the storage system anda readable storage medium, which have the above beneficial effects.

Description

technical field [0001] The present application relates to the technical field of data storage, and in particular to a storage system write cache data distribution method, device, equipment and a readable storage medium. Background technique [0002] Metadata transaction log synchronization means that during the IO process of the storage system, the metadata that has been generated will be used for transaction operations, and transaction logs between dual controllers will be synchronized. When a power failure or other failure occurs, the transaction log will be Persist to the disk. When the controller is powered on or restored, the transaction log information will be reloaded from the disk to redo the transaction log, thereby ensuring the data consistency of the storage system. After the metadata transaction log is synchronized, the synchronized request is sent to the metadata write cache, and the write cache module stores metadata to reduce the data response delay of the hos...

AI Technical Summary

Problems solved by technology

[0003] Under the traditional method, the transaction log module sends the data to the write cache module immediately after receiving the metadata and the synchronization is completed. However, due to the high data processing pressure of the write cache module, the write cache module often has too much data to be cached. In some cases, when the write cache cannot handle it, it will continue to cause a backlog of data, which will eventually cause the IO timeout of the upper host and affect the user experience.

Images