Distributed storage cache reading and writing method

Abstract

The invention discloses a distributed storage cache reading and writing method, which comprises the following steps of: receiving data objects to be written, allocating a cache space to each data object, and writing the data objects into the corresponding cache spaces; generating a corresponding data object fingerprint according to the data object, and allocating a storage node to the data objectfingerprint through the distributed hash table; when the storage node contains the current storage node, writing the data object into a disk of the current storage node; when the storage node does notcontain the current storage node, setting a local cache identifier for the data object fingerprint, and sending the data object to a remote storage node; by using the distributed caching method provided by the invention, a cache read-write unified entry is realized, the software complexity is low, the access time of the distributed storage system is reduced, and the system performance is improved.

Description

【Technical field】 [0001] The invention belongs to the technical field of computer storage, and in particular relates to a method for reading and writing a distributed storage cache. 【Background technique】 [0002] The storage system is developing in the direction of large-scale, distributed, and virtualized. Distributed storage systems have emerged as the times require. In distributed storage systems, cache systems have been introduced. Cache is defined as the system to process future data faster The storage capacity of the data retained by the request. Data stored in the cache is stored there because it is frequently accessed, or is a duplicate copy of data stored in another location. Caching essentially makes future access faster. [0003] The cache system of the current distributed storage system is mostly an intra-node cache, and business data is sent to the node through DHT or other discrete algorithms, and the node performs corresponding process processing after rece...

AI Technical Summary

Problems solved by technology

[0005] 2) The cache system will not cache the data of other nodes. For reading hot data, cross-node access is required, and the read performance is greatly affected.

[0006] 3) The read cache and write cache of the cache system are independent, which increases the complexity of the software; when reading data, it is necessary to read and write the cache first, and then read the read cache if it is unsuccessful, and then read the SSD POOL if it is unsuccessful. The path for reading is complicated. Multiple steps affect read performance

[0007] 4) Using SSD WAL cache as a persistence guarantee, the entire write performance is limited by the performance of SSD, and the write performance of the system is greatly affected

[0008] 5) The general-purpose distributed storage cache system is different for storage media such as HDD and SSD, and a set of solutions cannot be uniformly used

[0009] Due to the existence of the above defects, the data access time in the existing distributed storage system is long and the system performance is low

Images