Uygulama performansını artırmak için dinamik hafıza planlaması

In line with the increasing information processing requests of users and systems, the systems with much more powerful processing capabilities are needed. Despite the increase in hardware?s processing and transmission busses? speed, performance enhancement of the overall computing system may require simultaneous resource allocation depending on the computing task. To enhance the overall processing performance, research on software control and dynamic resource allocation of the computing and storage units have been underway by almost all the leading computer and computing nodes? manufacturers. In this study, computing system performance enhancement by dynamic memory scheduling has been developed.A fairly special computing process has been elaborated, the requested computing jobs are always created by choosing the inputs among a finite set inquiring the risk of the insurance policy of the candidate. In some special computing areas, like insurance risk investigation, calculations of income and premium need heavy and repetitive actuarial calculations, that the computing systems performing these computing efforts, may enhance CPU?s utilization by classifying on an ?intelligent manner? the computing jobs and caching the repeating computing jobs created by the same input interval. In this study, caching the repetitive jobs? results and enhancing the CPU usage rate has been presented and an intelligent regulation scheme has been introduced to software control.The jobs can be classified as two kinds; I/O jobs and computational jobs. External storage jobs are may be the examples of the I/O jobs. The computational jobs are the jobs, which always produce the same outputs versus the same input values. In this work, in the situation that the jobs are processed again; the advantages are examined to keep the outputs in the memory instead of reprocessing them.Instead of a static structure, to keep the process in a dynamic way to provide more benefit from the memory has investigated, the Iterative Learning Control methodology has been implemented to improve the usage of the system resources. Moreover, theoretical results, which obtained in this work has observed on the actual practice. The data that has been obtained from an individual retirement company has been corrupted by a linear function.