Improved edge cache replacement method considering partition

Abstract

The invention discloses an improved edge cache replacement method considering partition and belongs to the technical field of wireless communication. The method comprises steps of first dividing a cache region C into two regions, namely a new resource cache region C1 and an old resource cache region C2; estimating the popularity of a new video before the user accesses the video, and updating the actual popularity of the video when the video in the C1 and C2 is accessed every time; finally replacing the video in the cache region, removing the video whose cache time exceeds Ts in the C1, and preferentially replacing the video whose actual popularity is lowest in the cache region C; if the video content is not in the cache region C and is higher the lowest popularity of the cache region C andthere is no remaining space for storage, sequentially replacing files of the lowest actual popularity in the cache region until the storage space is enough. The method enables cell base stations to have optimal cache benefits, and adapts to rapid changes in the popularity of network access resources, thereby making the network performance more stable.

Description

Technical field [0001] The present invention belongs to the field of wireless communication technology, and specifically relates to an improved method for replacing edge buffers in consideration of partitions. Background technique [0002] According to the forecast of the Cisco Virtual Network Index, various forms of multimedia traffic will reach 90% of network traffic in 2018 (Reference 1: Cisco, "Cisco visual networking index: Global mobile datatraffic forecast update, 2012-2017," White Paper ,[Online]http: / / goo.gl / uQ0DJQ,2013.), with the rapid increase in the uplink and downlink rates of wireless access, the globalization of smart terminals, and the continuous development of self-media in the global environment, online video traffic is present The trend of exponential growth. Although the wireless communication technology has developed very rapidly in recent years, today's 4G access network can provide mobile users and stationary users with bandwidths up to 100Mbps and 1Gbps,...

AI Technical Summary

Problems solved by technology

[0006] On the one hand, the existing replacement algorithm has its own limitations. For example, the LRU algorithm considers replacing the least recently used content, which is very susceptible to interference from access noise. This interference is quite serious, resulting in unsatisfactory performance of the algorithm.

The LFU algorithm, which considers replacing content with the minimum frequency for a long period of time, also has flaws. This algorithm cannot promptly eliminate those content that have surged in visits in a short period of time in the past and have no cache value afterwards.

The first-in-first-out FIFO replacement algorithm will not only waste resources due to frequent replacement, but also cannot prolong the storage time of highly popular resources in the cache area, resulting in a low cache hit rate

In addition, some cache replacement strategies increase the number of requests served by cached content by extending the time certain content is in the cache space, which does not consistently guarantee that the cached value of those content resources that are cached for an extended period of time is higher than that of those that are not cached content resources for

In short, in addition to the defects of the existing cache replacement algorithm, most of the schemes do not start from the user's point of view, do not consider the user's behavior characteristics, and ignore the differences in the popularity trends of different types of network content. influences

Images