Algorithmic approach to caching strategy selection in web applications: a balance between performance, data consistency, and resource efficiency

Abstract

In modern web applications, which operate in environments characterised by high volumes of user requests and significant load variability, ensuring stable performance and efficient resource utilisation is of crucial importance. One of the key mechanisms for optimising performance is caching, which reduces the number of database accesses, shortens query processing times and reduces the overall load on the server infrastructure. However, choosing an optimal caching strategy is a complex task that requires consideration of numerous factors, including data update frequency, the nature of queries, memory constraints, and requirements regarding the timeliness of information. This article examines algorithmic approaches to selecting caching strategies in web applications and analyses the impact of cache parameters, including time-to-live (TTL), invalidation frequency and data access structure, on system performance. A generalised model for evaluating caching efficiency has been developed, taking into account load, latency and resource constraints. It has been shown that the use of adaptive caching approaches enables an optimal balance to be achieved between performance, data consistency and resource efficiency. The practical significance of these findings lies in their applicability to the design of modern web systems focused on high performance.