WhenLive streaming software constructionWhen the database access pressure is increasing, we can start to optimize from two aspects, the first is to improve the performance of the database itself, and the second is to introduce a cache to share the access pressure. Since improving the performance of the database itself is not unlimited, it is relatively practical to introduce caching.
Introducing a cache during the construction of live streaming software can share a certain amount of access traffic, effectively alleviate the pressure of database processing, and help the live streaming software achieve better performance. In order to get the most out of the cache, there are many issues that need to be paid attention to when using the cache.
Since the storage space of the cache is limited, it is necessary to eliminate part of the data through a certain data elimination strategy to store the newly arrived data in the cache built by the live broadcast software
1. First-in-first-out strategy
It mainly refers to the elimination of the first data stored in the cache.
2. The strategy has not been used for the longest time recently
It mainly refers to the priority of eliminating the most unused data in the cache, so as to ensure that the cache built by the live broadcast software stores all hot data and improve the cache hit rate.
3. Least frequent use of the strategy
It mainly refers to prioritizing the elimination of data that has been used the least number of times in a period of time.
In the cache built by the live broadcast software, the utilization rate of the cache is mainly determined by the cache hit ratio, which refers to the ratio of the requests that get a response through the cache to the total request, and the higher the cache hit ratio, the higher the cache utilization.
When building live broadcast software, the cache type can be divided according to the location of the cache, which can be roughly divided into the following types:
1. Database caching
It mainly refers to the query caching mechanism in the database.
2. Distributed caching
The distributed cache built by the live streaming software needs to be deployed separately, and hardware resources need to be allocated according to the requirements.
3. Local cache
It mainly refers to the cache implemented by using the server's local memory, because the server can directly read the cache in the local memory, so the data reading speed of the cache is relatively fast.
4. Reverse**
If the data is cached in the reverse, the response can be obtained directly from the reverse, thereby speeding up the response of the system.
As one of the important means to optimize system performance, caching is also indispensable in the construction of live broadcast software, however, caching will also cause some problems when using, such as cache avalanche, cache penetration, etc., so while introducing the caching mechanism, it is also necessary to do a good job in preventing related problems.
Statement: This article is original by Clouded Leopard Technology, **Please indicate the author's name and the original link, otherwise it will be regarded as infringement.