Arcadia Drives Healthcare Application Modernization and Cuts Costs Using AWS Enterprise Support

For nearly 20 years, Arcadia.io has helped healthcare organizations of all sizes reduce medical expenses, enhance risk-coding accuracy, and improve quality of care and patient health outcomes. The healthcare data and software company works with health systems, providers, payers, and life science organizations, and it offers integration technology to more than 50 physical and behavioral electronic health record (EHR) vendors. “We’ve helped our customers drive more than $2 billion in savings by avoiding medical expenses through more appropriately aligned, high-quality care,” says Jonathan Cook, chief technology officer for Arcadia.

An Amazon Web Services (AWS) customer since 2016, Arcadia runs its big data platform on Amazon Elastic Compute Cloud (Amazon EC2). The platform collects, aggregates, and normalizes data from 2,000 active interfaces each day that exchange data from EHRs, health systems, health plans, and lab vendors. The data flows into a data lake based on Amazon Simple Storage Service (Amazon S3). “We turn this information into insights for our customers to help them understand how to provide better care, reduce overall expenses, and make patients healthier,” Cook says. The company’s global dataset aggregates records from over 100 million patients.

Arcadia’s global dataset is growing at a rate of 40 percent annually, fueled by both new customer acquisition and an increase in the number of datasets that existing customers require to run their business. To scale its big data platform and data lake more cost-effectively, Arcadia sought to modernize its Windows-based application environment. “We wanted to reduce some of our operational costs, but we needed assistance in figuring out the best way to do that,” says Cook.

To achieve its business goals, Arcadia engaged AWS Enterprise Support for assistance with two key business initiatives. First, AWS Enterprise Support worked with Arcadia to fund a proof of concept (POC) and ultimately implement Amazon EC2 Spot Instances for the company’s daily patient data processing and analytics platform. “We knew AWS Enterprise Support could help us, and the team graciously offered to help fund the POC so we could find out whether this would work,” Cook says. AWS Enterprise Support also helped Arcadia use Spot.io APIs to supply Amazon EC2 instances and dynamically scale an Apache Spark big data cluster.

For the second initiative, AWS Enterprise Support helped Arcadia design and implement a Windows Server 2012 R2 upgrade for hundreds of Windows Server 2008 R2 servers running SQL Server. “Windows Server 2008 R2 was at end-of-life and we needed to upgrade,” Cook says. “We reached out once again to AWS Enterprise Support for help facilitating the upgrade with the least amount of downtime.” AWS Enterprise Support suggested using AWS Systems Manager, which provides a unified user interface for viewing operational data from multiple AWS services. By taking advantage of this service, Arcadia created a process for upgrading Amazon EC2 instances by replicating, upgrading, and then destroying old instances.

Working closely with the Technical Account Manager assigned to Arcadia as part of AWS Enterprise Support, Arcadia created a POC to modernize its application environment to run on Amazon EC2 Spot Instances using Spot.io’s technology for spot instance prediction and usage. “Working with AWS Enterprise Support, we tested the ability to use Amazon EC2 Spot Instances to automate and optimize our patient data environment,” says Cook. “As a result, we realized we would be able to reduce our operational costs by at least 60 percent. This gives us a lot of confidence that we can substantially lower our costs when we move to production.”

In moving to the latest version of Windows Server, Arcadia was able to update its systems in a way that did not affect its customers. “Relying on AWS Enterprise Support, we upgraded the majority of our Windows Server environment in one weekend, with no downtime,” Cook says. “This was a fast process, which took away the danger of trying to run an in-place upgrade, where you can sometimes have machines in between success and failure that can create downtime for customers. They rely on our solutions to serve patients and lower costs, so downtime is not an option.”

By communicating and collaborating frequently with AWS Enterprise Support through messaging tools, Arcadia receives technical support in minutes. “If we run into a problem and need to figure out why something is failing, it’s extremely helpful to have AWS Technical Account Managers who are able to quickly jump in and escalate the case to get it fixed,” says Cook.

As an example, Arcadia recently needed help determining why its Amazon S3 costs had risen more than expected. “Our AWS Technical Account Manager helped us realize we were being charged for writing tags on storage objects in Amazon S3,” says Cook. “We are doing this with millions of objects per week. After finding that out, we did a minor system tweak and reduced our costs by up to $20,000 a month.”

Arcadia is currently working with AWS Enterprise Support to drive additional modernization initiatives, such as implementing Amazon QuickSight to replace its current reporting and visualization tool. “In addition to Amazon QuickSight, we’re evaluating other AWS services such as Amazon Redshift for data analytics,” says Cook. “Working with AWS Enterprise Support has already brought business and technical benefits, and we look forward to uncovering more benefits in the future.”