Method, device and equipment for distributed storage client to read data

Abstract

The invention discloses a method for a distributed storage client to read data. The method comprises the steps that a reading request for reading the target data of a target node is received; according to the reading request, whether or not there is bitmap information corresponding to the target data in the cache of the target node is judged; if yes, the reading node corresponding to the target data is inquired according to the bitmap information, and the target data is read in the cache of the reading node; if not, the reading request is sent to a storage end, and the target data is read in the storage end. By means of the method, cache sharing between the nodes is achieved, the redundancy for reading the data is reduced, the memory utilization rate is improved, and meanwhile, the bandwidth waste is reduced. The invention further discloses a device and equipment for the distributed storage client to read the data and a computer readable storage medium. The device and equipment for thedistributed storage client to read the data and the computer readable storage medium have the above advantages.

Description

technical field [0001] The present invention relates to the field of distributed storage systems, in particular to a method for reading data by a distributed storage client, and also relates to a device, equipment and computer-readable storage medium for reading data by a distributed storage client. Background technique [0002] The distributed storage system is to disperse and store data on multiple independent devices through the network, so as to reduce the pressure on the network, and has high reliability and security. The client cache of distributed storage is a logical unit used by client nodes to temporarily store read and write data to improve read and write performance. Distributed storage connects storage nodes together to form a cluster. Each storage node has a corresponding cache space. When any node in the cluster reads data from the backend, it will temporarily store the read data in the corresponding in the cache. [0003] In the existing technology, storage...

AI Technical Summary

Problems solved by technology

That is to say, the data returned by each node to the front end can only come from the cache space of this node. In this case, even if there is target data in other nodes that provide the same external service, the target data cannot be used

Since each storage node independently maintains its own cache content, coupled with the inherent disadvantages of mechanical disk read performance, in this case, it will cause great read data redundancy, reduce memory usage, and waste bandwidth

Images