Komodo Health Expedites Onboarding, Enhances Efficiency Using AWS-Powered Data Analytics
The founders of Komodo Health, Web Sun and Dr. Arif Nathoo, envisioned a healthcare analytics solution that would improve healthcare outcomes and reduce the burden of disease by unifying and harnessing the fragmented medical data collected from millions of people throughout the world. They proposed using algorithms to process clinical knowledge and patient experiences over decades to produce meaningful insights. To process that amount of data, Komodo Health needed a computing solution that didn’t require constant management and that helped its engineers focus on what they do best: work with data.
To accomplish that, Komodo Health turned to Amazon Web Services (AWS) and used Amazon Elastic Kubernetes Service (Amazon EKS)—a fully managed Kubernetes service that runs critical applications with security, reliability, and scalability—and Amazon EMR, an industry-leading cloud big data service for processing vast amounts of data using open-source tools. “Before we made the migration to Amazon EMR, our data scientists were spending a lot more time on infrastructure,” says Chris Han, infrastructure team lead at Komodo Health. “But the old model couldn’t scale. We wanted to shift focus for the data scientists to start actually working with the data.” Implementing Amazon EMR and Amazon EKS did that and much more, solving several problems for Komodo Health and producing the benefits of decreased onboarding time, quick scaling, and decreased costs.
Now data scientists don’t have to work deeply with infrastructure. Using Amazon EMR, they can start working with the data on day one.”
Infrastructure Team Lead, Komodo Health
Migrating from “Old World” to “New World” Architecture
Founded in 2014, modern healthcare intelligence service Komodo Health has a mission to reduce the burden of disease. The company unifies data across fragmented, disparate sources, including 11 billion lab records, 15 million daily healthcare encounters, and decades of findings from clinical trials. By integrating this data, Komodo Health unlocks meaningful opportunities to close gaps in care by providing actionable insights for researchers, life sciences companies, health plans, and advocacy groups.
Komodo Health’s previous architecture required a lot of management and was not the optimal infrastructure for the massive amounts of data that Komodo Health was harnessing. “In the old world, a lot of our data scientists were self-managing their infrastructure, and it was a problem,” says Han. “It would take data scientists weeks to get ramped up and get onboarded to start working with our data.”
Adopting Amazon EMR brought Komodo Health into “a new world,” according to Han. “Now data scientists don’t have to work deeply with infrastructure,” says Han. “Using Amazon EMR, they can start working with the data on day one.” The implementation of Amazon EMR runs Apache Spark, making it a more managed, cost-effective system overall. Komodo Health also optimizes costs by using Amazon Simple Storage Service (Amazon S3), an object storage service that offers industry-leading scalability, data availability, security, and performance.
Using Amazon EKS to Run a Multitenant Notebook Solution
Komodo Health uses Amazon EKS to manage the workloads it runs using JupyterHub, a notebook solution from the open-source Project Jupyter. The company’s data scientists use the notebooks to explore data, create queries, and design algorithms while Amazon EMR performs complex computations on the backend.
Komodo Health’s use of Amazon EMR, Amazon EKS, and Amazon S3 provides several benefits. First, these services reduce the operational burden on engineers, who no longer need to run the computing infrastructure. Second, the system—built on these managed services—reduces the onboarding time for engineers and data scientists from weeks to minutes. Third, Komodo Health has seen a reduction in costs and an improvement in performance. Ultimately, Komodo Health uses AWS services to scale quickly, accelerating the overall engineering organization, including the company’s data scientists’ output. One other unexpected benefit has also revealed itself: because the data has become simpler to use, nontechnical departments are able to access the data. For example, Komodo Health’s clinical experts, customer success teams, and business analysts can now analyze data to drive insights that ultimately support patient care.
In pursuit of its mission, Komodo Health is using AWS to map real-world patient data and the complex ripple effects of the COVID-19 pandemic, tracking the decline in medical testing, cancer screening, and children’s immunizations. “We’ve gained the abilities to have more sophisticated rules embedded as part of our data processing and to process more complex rules on a larger scale and with more complex datasets. These abilities have helped us complete analyses more quickly and validate them iteratively,” says Doug Lawrence, vice president of engineering at Komodo Health. Producing and analyzing reliable data is essential to stopping the spread of the COVID-19 pandemic and managing the secondary effects of delayed care. Data analysis tasks that once took Komodo Health days to run now take only hours.
Breathing Life into Data and Healthcare
Using its powerful computing infrastructure backed by AWS, Komodo Health breathes life into data, supporting various segments of the industry in an effort to improve healthcare for millions of people. “We’ve been on a hypergrowth trajectory, and having a cloud provider like AWS certainly helps us do that without worrying about capacity planning or the availability of resources,” says Jean Barmash, vice president of engineering at Komodo Health. “On AWS, not only do we have the compute power, the networking, and the storage available to us, but we also have access to expertise.”
Komodo Health provides software for the healthcare industry, but its mission is not just to gather data, design applications, and crunch numbers—it’s also to close gaps in patient care. For example, Komodo Health’s Pulse solution identifies patients who may have rare diseases but are floating around the healthcare system in need of correct diagnoses. “Being able to look for key events across a patient’s entire history enables us to come up with an aggregate prediction of a certain disease,” says Barmash. “On top of that, we run additional enrichments to build that patient profile and spot red flag symptoms. Our combination of data makes the results really powerful.”
Pursuing its mission to reduce the burden of disease, Komodo Health harnesses the power of Amazon EMR and Amazon EKS to unify massive amounts of data. And using that data, Komodo Health can provide actionable insights for researchers and healthcare providers, ultimately improving the lives of millions.
About Komodo Health
Modern healthcare intelligence service Komodo Health transforms extraordinary amounts of data into rich and meaningful insights that can improve healthcare and reduce the burden of disease.
Benefits of AWS
- Reduced operational burden on engineers
- Reduced onboarding time for engineers and data scientists from weeks to minutes
- Runs data analysis tasks in hours instead of days
- Enables nontechnical departments to access and benefit from data
AWS Services Used
Amazon EMR is the industry-leading cloud big data platform for processing vast amounts of data using open source tools such as Apache Spark, Apache Hive, Apache HBase, Apache Flink, Apache Hudi, and Presto.
Amazon Elastic Kubernetes Service (Amazon EKS) gives you the flexibility to start, run, and scale Kubernetes applications on the AWS Cloud or on-premises.
Amazon Simple Storage Service (Amazon S3) is an object storage service that offers industry-leading scalability, data availability, security, and performance.
Companies of all sizes across all industries are transforming their businesses every day using AWS. Contact our experts and start your own AWS Cloud journey today.