Geisinger Health System is well known in the healthcare community as a pioneer in data and analytics. They were one of the first adopters of Electronic Health Record (EHR) in 1996 and went with Epic. In addition, they used an Enterprise Data Warehouse (EDW) since 2008. Much of the daily and weekly operational reporting, as well as an abundance of ad hoc analytics, come from the EDW. As you can imagine, having huge volumes of clinical as well as claims data, they wanted to find the best way to take advantage of the data and use it for insights. That’s where Hive 2 with LLAP steps in. Watch the video here.
Approximately 18 months ago, the Data Management team implemented Hadoop using Hortonworks Data Platform (HDP). Their successes in implementation and development have proven to the organization that they should abandon the traditional EDW in favor of the Hortonworks Data Platform.
In less than 18 months, they installed HDP, created a data ingestion pipeline, duplicated all source feeds from the EDW into HDP, and had several analytics developed with HDP and Tableau. Furthermore, they exploited the new capabilities of the platform, using Natural Language Processing (NLP) to interrogate valuable (but previously hidden) clinical notes. The new platform has data that is modeled and governed, setting the stage to push Geisinger Health System from a pioneer to a leader in Big Data and Analytics.
Furthermore, by upgrading immediately to HDP 2.6 when it was available in April 2017, they learned they could take advantage of Apache Hive 2 with LLAP, where they were able to get 20 of their 22 queries to perform in under 2 minutes.
Geisinger was one of the speakers at Hortonworks Data Summit. Listen to their video here. where they go into detail in specific use cases on transforming their EDW system into Hadoop. In one example, what use to take seven to eight hours loading data into their EDW, took only two and half hours using HDP. This included 7,200 tables from their EHR system. Read the full case study.