A Survey on Cognitive Storage by using Data Popularity in Big Data

Data amount is increasing rapidly, indicating that we are in the era of big data, which brings opportunities and challenges. Storage system efficiency can be significantly improved by determining the value of data. The computers can be taught to learn the difference between high value and low value data. The main concept is to make computer learn what to remember and what to forget. In storage systems, the relevance of �les to users can be taken into account to determine storage control policies to reduce cost, while retaining high reliability and performance. The relevance of a �le can be estimated by applying supervised learning and using the metadata as features. We introduced and redefined the data value, and proposed the heterogeneous storage architecture based on the data value, which can dynamically calculate the data value and choose the appropriate storage strategy according to the different data value and improve the overall performance of the system. Popularity prediction in microblogging network aims to predict the future popularity of a tweet based on the observation in the early stages. Existing studies have investigated many features for prediction.