Southern Illinois University MySQL Health Trend of Patients Discussion Part 1: As our object is to find the health trend of these patients, and the indica

Southern Illinois University MySQL Health Trend of Patients Discussion Part 1:

As our object is to find the health trend of these patients, and the indicators are blood sugar, pulse, and temperature, we have to collect and isolate these parameters from the data first. So the data should be ingested to Hadoop from MySQL.

The second stage is data processing, which will performed by Spark and MapReduce. The unstructured data and information is stored in distributed file system HDFS and the NoSQL distributed data HBase.

Next stage is analyzing. Data will be interpreted by the processing framework such as Hive and Impala.

The fourth stage is accessed, which is performed by a tool such as Hue and Cloudera search. We can utilize Hue’s Imapala to find the health trend of patients with those parameters.

After we have the results and interpretation of the data, we also have to consider the limitations in the process.

Part 2:

In this case we can find that health care software provider utilize mysql to deal with the problem.

However, sql only can figure out structure data. None structure like patient prescription can not be done by sql. On the other side, Apache Hadoop is a collection of open-source software utilities that facilitate using a network of many computers to solve problems involving massive amounts of data which contain structure data and none structure data. (Bappalige, 2014)

We can utilize it strength which is good at distributed storage and processing of big data to build model.

As a data analytics manager, we can use Hadoop to work in parallelize large-scale batch computations to process big data sets. In this caseHadoop Distributed File System (HDFS) – distributed , fault-tolerant storage system shows it advantages to petabytes of data.Hadoop MapReduce is a common technic in HDFS.

The health care trend of patients should focus on 3 indicator. Our task is to use the trend in the past to find build model and then set database system and help us query the plenty of data in a easy way. When we finish the model we can query by patient’s ID and selecting blood sugar indicator to forecast patient heath.