Comparative Research on Active Learning of Big Aata based on Mapreduce and Spark

Abstract The big data era is coming, and it is changing itself in science, engineering, medicine, healthcare, finance, business, and ultimately in our society. Traditional data may not be able to handle large amounts of analytical data. Big data can appear in three formats: structured, unstructured, and semi-structured. Map Reduce and Spark are the two most popular open-source important frameworks for large-scale data analysis. The performance of Map Reduce and Spark will vary depending on the application being implemented. The Map Reduce program model is a Hadoop configuration used to store large data formats stored in the Hadoop File System (HDFS) in big data. This is an integral part of the Hadoop framework that contains its core parts. Map Reduce has failed those programs for real-time data processing as it is designed to do volume processing on large volumes of data. Apache Spark is a data processing framework that can quickly handle tasks on large datasets, and whether on its own or in series with other distributed computing tools, across multiple computers and data processing tasks can be distributed. Apache Spark is a data processing framework that can quickly handle tasks on large datasets. Whether on its own or in series with other distributed computing tools, across multiple computers and data processing tasks can be distributed. The Map Reduce is two times larger than the spark. Thus, the Spark model quickly deploys and Apache spark, proving that data is more effective than a Map Reduce.