ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 4. Vol. 25. 2019

DOI: 10.17587/it.25.216-222

D. R. Potapov, Postgraduate, e-mail: potapovd36@gmail.com, Voronezh State University, Voronezh

Cache Efficiency Research for Using in Adaptive Data Storage

There are a lot of applications with a large amount of the static data or data which is using for reading mostly. At this rate cache applying improves performance greatly. In this case an adaptive data storage can change cache size during execution to achieve maximum efficiency. There are two main factors which affect container performance. The first one is a set of requesting data, which in common case can be described as Gaussian distribution. The second is a difference between speed of a main container and the cache. This paper provides a research results of a cache efficiency depending on the standard deviation of normal distribution and storage speed coefficient. The survey takes into consideration combinations of several parameters. One of such parameters is a data structure. The research is carried out for different programming language C# data structures (Dictionary, SortedDictionary and SortedList). Another important parameter is a cache replacement policy. Various cache algorithms (Least Recently Used and Most Recently Used) are also examined in this research. Adaptive storage performance is evaluated as sum of a search time in the main container and a search time in the cache. The research results are analyzed in the article and the main patterns for calculating the optimal cache size for using in adaptive data storage are identified.
Keywords: store the data, cache efficiency, optimal data storage, normal distribution, adaptive data container, optimal cache size, data structures, container load, cache algorithms, dynamic cache

P. 216–222

To the contents