Abstract: In this big data world we can see daily based data which are generated day by day e.g. Google, amazon, Facebook etc. This large amount of data is un-reliable to store, manage and analyze. This large volume of data runs on Commodity hardware, which are parallel arranged. The data that are large in volume takes more time to execute for particular process and causes fail to distributed system because these huge volume data runs on distributed system. To overcome this issue a framework was proposed called HADOOP for big data processing and is being used now days in organizations. It process large amount of data in small amount time rather than large distributed systems. It has automatic fault tolerance capacity to tackle with failed of systems in execution or processing time using Map Reduce programming technique. Here execution time is still an issue for delivering large amount of data and processing it repeatedly for particular oracle. Existing HADOOP system does not have any feature to reduce time for recompilation. I propose a HADOOP distributed system for large data processing, which has two type of cache memory one is local cache, and other is distributed centralized cache memory. I use these cache memories for reducing the recompilation time. Distributed cache consisted with least memory space so LRU cache is implemented to allocate memory space simultaneously. The LRU caching scheme is to remove the least recently used frame when the cache is full and a new page is referenced which is not there in cache.
Keywords: big data, distributed, Map Reduce, HADOOP, LRU cache.
Title: DISTRIBUTED DATA CACHE FOR BIGDATA WITH LRU CACHE USING MAPREDUCE TECHNIQUE
Author: R.Saranya, N.Sundaram
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
