Online Configuration of NoSQL Databases for Optimal Performance of Time-Varying Workloads

Back to all technologies
Download as PDF
2019-BAGC-68646
Innovators at Purdue University developed a system known as Sophia to reconfigure a NoSQL Database Management System for highest throughput as the application characteristics change.
Sophia increases the economic lifecycle of databases through a Cost Benefit Analysis (CBA) coupled with a Reconfiguration Plan, implements intelligent incremental reconfigurations, and predicts future workloads up to an hour ahead using a Workload Predictor, thereby optimizing throughput. In addition, the Sophia program will initiate reconfiguration only when it estimates the benefit over a future time window is higher than the cost of the reconfiguration. The invention is efficient for cloud computing as well as creating and managing on-premises databases.
Benefit

Advantages:

-Adaptable for changing workloads, including no change
-Compatible with a variety of NoSQL databases
-Keeps data available to users during reconfiguration
-Performs Cost Benefit Analysis to determine optimal reconfiguration

Potential Applications:

-Cloud computing
-Distributed NoSQL databases
-Data analytics for scientific research
Apr 16, 2020
Utility-Gov. Funding
United States
(None)
(None)

Jul 5, 2019
Provisional-Patent
United States
(None)
(None)
Purdue Office of Technology Commercialization
The Convergence Center
101 Foundry Drive, Suite 2500
West Lafayette, IN 47906

Phone: (765) 588-3475
Fax: (765) 463-3486
Email: otcip@prf.org