NEW! Mirantis Academy -   Learn confidently with expert guidance and On-demand content.   Learn More

CASE STUDY:

GlaxoSmithKline Accelerates Drug Discovery with Docker Enterprise

Leveraging containers to accelerate drug discovery at GSK


The following excerpt is from a post originally published in 2017 on the Docker blog.

Leveraging Data Science for Improved Outcomes

The biggest challenge in pharmaceutical research is that hundreds of drug formulations need to be created to take one successfully to market, because only 3% of formulated molecules actually become medicine. Lindsay Edwards heads a Data Science group that is focused on respiratory illnesses such as Chronic Obstructive Pulmonary Disease (COPD) and asthma. His group uses big data analytics to mine research data and previous patient trial data to arrive at results more rapidly.

Lindsay needed a faster way for his team to safely test and experiment these new technologies across different hardware platforms, while also allowing his scientists to easily share their research amongst each other. His organization needed an agile platform that could support different software tools, applications, and hardware configurations, and still be able to scale these tools up as needed.

Rapid Prototyping with Edge Node On Demand

To meet the needs of the data science group, Ranjith Raghunath, the director of Big Data Solutions, needed a way to quickly deliver new technology stacks to various researchers, independent of the underlying infrastructure. He explored various options and selected Docker Enterprise because it is the most effective platform for delivering secure but isolated environments for researchers to use.

These environments come preconfigured with necessary enterprise integrations such as authentication through Active Directory, but researchers can also install and work with their preferred software components. Using Docker Enterprise allows for complete infrastructure independence and true app portability, enabling Lindsay’s team to move from one cluster to another while keeping everything they have intact.

The new solution, named Edge Node On Demand, is powered by Docker Enterprise and integrates with various Cloudera clusters and MongoDB. The solution provides GlaxoSmithKline these key benefits:

  1. Single interface– Standardizing on Docker Enterprise means all the different apps can be treated in a single consistent manner. Further, Ranjith’s team leverages the Docker APIs to consolidate environment information to a single interface for both service requests and tracking application deployments.

  2. User isolation– Each researcher has a sandbox for experimentation that is isolated from others, so they can break their own experiments without affecting others.

  3. Reusability– If a researcher does make a discovery, using Docker Enterprise makes it easy to rebuild and redeploy the same application again and again.

  4. Seamless migration– Whether it is moving from a dev environment to production or from one hardware configuration to another, Edge Node on Demand is truly portable across environments for seamless migration.

  5. Sharing– Docker Enterprise makes it possible for GlaxoSmithKline to share research and data easily, improving collaboration and accelerating results.

video thumbnail

GSK Leverages Docker Enterprise to Improve Research and Accelerate Drug Discovery

Challenge

Prototyping new data science tools took 3 to 6 months, slowing down drug discovery and clinical trials.

Solutions

Docker Enterprise is the cornerstone of a “Edge Node On Demand” platform that secures and isolates research environments which include Cloudera and MongoDB. Each researcher gets an isolated, secure sandbox.

Results

Faster prototyping of analytics applications and the underlying technology stack allows scientists both inside and outside GSK to work more efficiently.

GSK By the Numbers

£30

Billion in Revenue

£3.9

Billion Invested in R&D

100,000

Employees

“Docker Enterprise allows GSK to support a multitude of tools and technologies and interfaces so that scientists can run data analysis at scale.”

— Ranjith Raghunath

Director of Big Data Solutions

Additional Case Studies