Implementation of caching algorithm in scady Grid Framework

Abstract

Demand for High Performance Applications has already grown in present scenario. Grid Computing and Cloud computing are the requirement of today and tomorrow. Many UNIX based and command based Grids are already in use. But people generally prefer window based machines. Considering this situation, we created a Grid framework called SCADY (Scalable & Dynamic) which is based on Alchemi.Net Grid Framework. From the literature review and findings it has been observed that a better fault tolerance mechanism is needed in these grid frameworks to ensure execution efficiency. This paper proposes a caching algorithm for better fault tolerance in Scady Grid Framework. The proposed algorithm uses the cache write-through method in which at the same time the data is written into the cache block and the corresponding main memory location. The data cached can be retrieved very fast on demand, while it is ensured that nothing will get lost if a crash, power failure, or other system disruption occurs as the same data is in main memory. Write through method is the most preferred method for storing the data for the applications where data is critical and data loss cannot be tolerated. The algorithm is implemented and tested for efficiency. It is found that it gives comparatively better result.