Proceedings of the International Conference , “Computational Systems for Health & Sustainability” 17-18, April, 2015 - by R.V.College of Engineering, Bangalore,Karnataka,PIN-560059,INDIA All Copyrights Reserved by R.V. College of Engineering, Bangalore, Karnataka Page | 179 Best Practices Generation for Storage Area Network: A Review Sudarshini Tyagi B.E., 2 nd year, Department of Computer Science and Engineering R.V.C.E, Bangalore A.Kowcika Assistant Professor Department of Computer Science and Engineering R.V.C.E, Bangalore Abstract: This paper presents a survey on SPIKE, an automated best practice generation tool that uses a combination of information retrieval principles, entity ranking and decision-tree classification to statistically infer the best practices for SAN configuration problems and a system which is designed to generate best practices using ILP takes a relational system management repository and the set of problem reports rose against the same managed repository as input. The output of such a system includes a set of best practices with confidence values. And a comparative analysis on SPIKE and ILP is highlighted. Keywords: Best Practices, Decision tree learning, ILP, SAN. 1. Introduction 1.1. Machine Learning Machine Learning is a type of Artificial Intelligence exhibited by Machines. It provides machines, the ability to learn without actually a human teaching it i.e. we don’t need to program the machine. Rather, we create such algorithms which help it to learn from its past experience.It focuses on the development of the computer programs as they are exposed to new data. Machine Learning is usually muddled with Data Mining. It is primarily due to the similarities between them. Both the systems look through the data to search for the patterns. However, there exists a basic difference between the two of them. In Data Mining, the patterns recognized are used for human reference. On the other hand, in machine learning, the detected patterns are used to bring changes in the program. This makes the system intelligent. For example, Facebook's News Feed changes according to the user's personal interactions with other users. If a user frequently tags a friend in photos, writes on his wall or "likes" his/her links, the News Feed will show more of that friend's activity in the user's News Feed due to presumed closeness. Another very good application of Machine Learning can be observed on websites like Quora and Flipkart. Here, on Quora, user activities like up voting/down voting an answer, commenting on the answers and following particular topics act as input data to the machine learning algorithm. All this data is then analysed and the detected pattern is used to change the program, hence changing the feed on your timeline. This makes the social-networking sites user-friendly as they get to see what they want to see more without particularly searching for the required topics. Flipkart also uses Machine learning techniques in order to display the current trends based on the customers’ feedback and the number of purchases of a particular product. Other applications of Machine Learning include Spam Filtering, optical character recognition (OCR), search enginesand computer vision.