Data loading method and device

Abstract

The invention discloses a data loading method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the following steps: receiving a data loading request, and determining that the state of a key name in the data loading request is a common state; translating the key name according to the current offset, splicing the translated key name with the current offset, generating a new key name corresponding to the key name, updating the current offset, then judging whether the new key name exists in the cache or not, and if the new key name exists in the cache and the key value type corresponding to the new key name is an identification code type, waiting for a set time, repeating the step until no new key name exists in the cache; locking the current key name in the cache, and generating an identification code as a key value corresponding to the current key name; and acquiring data corresponding to the key name from the database, determining that the identification code is valid, and updating the current key name and the key value in the cache. According to the embodiment, other requests are prevented from being blocked while repeated loading is avoided, and the data loading efficiency is improved.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a data loading method and device. Background technique [0002] In the process of using Redis cache, two types of problems will be involved, one is to cache singleton loading, and the other is to avoid hot key (key name) problems. Among them, the hot key problem refers to the problem that due to a large number of requests accessing the same key at the same time, the pressure on the single node of Redis is too large, which leads to the problem of cache collapse. [0003] In order to implement caching singleton loading, existing technologies usually implement distributed locks. In order to eliminate the impact of hot keys on Redis, the existing technology usually adds random values ​​after the key, so that a key has multiple backups on different nodes to avoid single-node pressure; local cache can also be increased to give priority to local access when accessing Cache, thereby ...

AI Technical Summary

Problems solved by technology

[0005] Use distributed locks to control the singleton loading of the cache. When the request to acquire the lock is interrupted unexpectedly for some reason, the acquired lock needs to wait for a certain timeout period before it can be released, resulting in all subsequent requests being blocked. The query efficiency is affected; the use of random values ​​​​eliminates the problem of hot keys, and the cached data cannot be deleted accurately; the use of local caches to eliminate the problem of hot keys results in high system complexity

Images