SURVEY ON URGENCY DATA PARTITIONING IN MEDICAL SERVICES USING ABUNDANT DATA ON HADOOP CLUSTERS

Keywords

Hadoop, Map reduce, Internet of things, Frequent Item set Mining, Data Partitioning

Abstract

Big data is large volume of data for both structured and unstructured data that includes a business on a day to day basic. Cloud computing is popularizing the computing paradigm in which data is outsourced to a third-party service provider (server) for big data. The Internet of Things (IoT) is the interconnection of uniquely identifiable embedded computing devices within the existing internet infrastructure. Delivering clinical information of patient at the point-of-care to physicians is critical to increase the quality of healthcare services, especially in emergency time. However, clinical data are distributed in different hospitals. It is sometimes difficult to collect clinical data of patient ubiquitously in case of urgency. In order to support the ubiquitous content accessing a resource model is first proposed to locate and get clinical data which are stored in heterogeneous hospital information systems using HDFS (Hadoop Distributed File System).In the proposed method clinical data of patient is defined as resource with unique URL address. Related clinical data of one patient is collected together to form a combinational resource, and could be accessed by physician if authority is assigned to the physician, by using a mySql database technique efficiently in big data applications for better performance and scalability. This type of database support faster execution of queries compared to non-relational databases. By implementing the system that combines IoT with Big Data is built to provide quick and effective for different patients. We implement frequent data process on a 24-node Hadoop cluster, driven by a wide range of datasets created by Quest Market-Basket Synthetic Data Generator.