No abstract

Abstract. With the rapid growth in technology, there is a huge proliferation of data in cyberspace for its efficient management and minimizing the proliferation issues. Distributed file system plays a crucial role in the management of cloud storage which is distributed among the various servers. Many times some of these servers get overloaded for handling the client requests and others re-main idle. Huge number of client requests on a particular storage server may in-crease the load of the servers and will lead to slow down of that server or dis-card the client requests if not attended timely. This scenario degrades the over-all systems performance and increases the response time. In this paper, we have proposed an approach that balances the load of storage servers and effectively utilizes the server capabilities and resources. From our experimental results and performance comparison of proposed algorithm with least loaded algorithm we can conclude that our approach balances the load, efficiently utilize the server capabilities and leverage the overall system performance.

Mining information from a database is the main aim of data mining since years. The most relevant information which one requires as a result of data mining is getting associations between various attributes. More preciously mining frequent itemset is the most significant step to initiate the mining operation. Most of the algorithms discussed in the literature require multiple scan of the database to get the information on various sub steps of the algorithm which becomes quite computationally extensive. In this paper, we are proposing an algorithm Lexicographic Frequent Itemset Generation (LFIG), which can extract maximum information from a database only in one scan. We will use Lexicographic ordering of attributes and arrange itemsets in multiple hashes which are linked to their logical predecessor.

No abstract

This is a special issue as an outcome of 'RAPSCON-2019' sponsored by APTI and organized by Sri