Hybrid- and Multi-Cloud Storage Strategies for Cost-effective Deployment of Data Resources

PROJECT SUMMARY The goal of the FaceBase III Hub, supported by the National Institute for Dental and Craniofacial Research (NIDCR), was to create a data resource to serve the entire community of dental and craniofacial researchers first by sharing diverse data related to craniofacial development and

dysmorphology and increasingly dental and translational research, as well as other research that can benefit from the data that is in the FaceBase repository. FaceBase is home to over 1,040 imaging, genomics and other datasets, spanning human, mouse, zebrafish, chick with prospective contributors in other species, and it has seen steady usage worldwide.

FaceBase is in the process of migrating from an on-premise infrastructure to the Amazon Web Services (AWS) “cloud” service provider. While the cloud deployment of FaceBase will offer many benefits to the community that it serves, one particular challenge facing FaceBase or any other data resource operated on a commercial cloud is the incremental and unpredictable costs

of data storage and downloads. In this project, we propose to explore hybrid- and multi-cloud storage strategies in order to optimize for the most cost-effective storage providers and schemes available. We propose the following specific aims: (1) leverage cost competitive specialized cloud storage service providers

to supplement core service deployment on AWS, and (2) explore the usage of on-premises storage resources through a hybrid-cloud strategy to augment further our public cloud infrastructure and seamlessly exploit subsidized low-cost storage infrastructure through our institution. The ultimate goals of the project are to accelerate dental, oral, and craniofacial research and to

advance the understanding of cost-effective utilization of cloud resources to benefit potentially any NIH supported data resources. PREVIEW Date: Apr 11, 2023 Workspace ID: WS01078345 Funding Opportunity Number: PA-20-272