Method for solving multi-process access to Level DB based on RPC proxy

Abstract

The invention relates to the field of access of an RPC (Remote Procedure Call) to a LevelDB (Level DB), in particular to a method for solving multi-process access to the LevelDB based on an RPC proxy, which comprises the following steps of: S1, starting an RPC proxy server side module at a LevelDB server side, and finishing subsequent process read-write operation; and S2, at the LevelDB client, starting an RPC proxy client module, automatically connecting the RPC proxy client module to an RPC proxy server module through an RPC connection module, and completing subsequent process read-write operation after the RPC proxy client module is ready. According to the invention, based on the server side and the client side of the RPC, a plurality of local and remote processes can simultaneously initiate various read-write requests to the LevelDB, so that greater use flexibility and convenience can be brought to the LevelDB, and meanwhile, a large amount of unnecessary storage space waste is avoided.

Description

technical field [0001] The invention relates to the field of RPC access to LevelDB, in particular to a method for solving multi-process access to LevelDB based on RPC proxy. Background technique [0002] LevelDb is a C++ library capable of handling billion-level key-value data persistent storage. LevelDB is developed by Google and is open source. Well-known projects such as Chrome browser and the bottom layer of IPFS file system use LevelDB as metadata management engine. As LevelDB is widely used, an obvious defect limits its further development. LevelDB only allows single-process access. If you want to achieve concurrency, you can only achieve concurrency through multi-threading under a single process. In reality, there are Two obvious usage scenarios: one process reads and writes LevelDB, and other processes perform read-only operations on LevelDB; one process reads and writes LevelDB, and other local processes and even remote processes also read and write the same LevelD...

AI Technical Summary

Problems solved by technology

As LevelDB is widely used, an obvious defect limits its further development. LevelDB only allows single-process access. If you want to achieve concurrency, you can only achieve concurrency through multi-threading under a single process. In reality, there are Two obvious usage scenarios: one process reads and writes LevelDB, and other processes perform read-only operations on LevelDB; one process reads and writes LevelDB, and other local processes and even remote processes also read and write the same LevelDB

[0003] In these two scenarios, LevelDB does not provide an ideal solution; the first problem is a roundabout solution. For other read-only LevelDB processes, it is necessary to re-copy LevelDB data. The disadvantage is that the storage space is not good. Necessary waste, especially when the amount of data reaches TB or even PB level and there are multiple read-only processes

The second problem has no mature solution at all

Images