R&D, Optimization on HPC

• Optimization and multi-cloud orchestration engine
• Powered by Machine Learning and in-memory database technology on High Performance Computational (HPC) Engine
• Allow models to be run at scale (millions of virtual patients) and reduce lead time significantly 

qPar is unique in the industry for QSP and other use cases:

• Running up to tens of thousands of tasks in parallel from a single qPar node
• Reliably manage orchestration and scheduling of the tasks
• Utilize AWS spot instances delivering over 50% cost savings over running on-demand
• Capture task failures including stack trace from remote nodes
• Capture all console output from all the tasks and view in Matlab
• Very simple to administer - all tasks run on the same image maintained by RxDS
• Efficient: Instances starts within 1-3 minutes, tasks start within 5 minutes all in parallel
• Simple API for the Matlab or Julio programmer supporting remote calls between both
• Support multiple client-side technologies (eg. Julia, Matlab, Python, C, C++, Javascript, etc.)

*Kx and Kdb+ are registered trademarks of Kx Systems, Inc. MATLAB is a registered trademark of The MathWorks, Inc. The use of these modules require purchase of appropriate MATLAB and KDB+ licenses from the respective companies.