Framework de cache pesquisável e de alta performance

Abstract

Nowadays, all organizations need to share and store information for future reference, and the larger the organization is the greater will be the flow of data and the number of persons accessing it concurrently. It is through a central databases that is possible to keep all this information consistent and accessible whenever needed. When a organization grows to the point of saturating the bandwidth or processing power of their database, it is necessary to think of innovative ways to keep their quality of service. Usually most system bottlenecks are located in the database access, being this one of the most important places to keep as efficient as possible. One approach that recently is gaining some momentum and that in most situations its enough, is by using in-memory caching systems to temporarily store the most frequently used data. However if there is the need to store a data-set (or part of it), there is no way to make a search in this data because this systems typically have a key-store architecture, similar to a dictionary. The current alternatives are the use of hybrid databases, which place in memory the current data-set, but still need a slower storage medium. The work done in this thesis had the objective to research and determine whether it would be possible to implement a high efficiency and searchable in-memory caching framework while maintaining the characteristics of an in-memory caching system.