A web cache replacement strategy for safety-critical systems

Abstract

A Safety-Critical System (SCS), such as a spacecraft, is usually a complex system. It produces a large amount of test data during a comprehensive testing process. The large amount of data is often managed by a comprehensive test data query system. The primary factor affecting the management experience of a comprehensive test data query system is the performance of querying the test data. It is a big challenge to manage and maintain the huge and complex testing data.To address this challenge, a web cache replacement algorithm which can effectively improve the query performance and reduce the network latency is needed. However, a general-purpose web cache replacement algorithm usually cannot be directly applied to this type of system due to the low hit rate and low byte hit rate. In order to improve the hit rate and byte hit rate, a data stream mining technology is introduced, and a new web cache algorithm GDSF-DST (Greedy Dual-Size Frequency with Data Stream Technology) for the Safety-Critical System (SCS) is proposed based on the original GDSF algorithm. The experimental results show that compared with state of the art traditional algorithms, GDSF-DST achieves competitive performance and improves the hit rate and byte hit rate by about 20%.