학술논문

Cross-institutional research cyberinfrastructure for data intensive science
Document Type
Conference
Source
2016 IEEE High Performance Extreme Computing Conference (HPEC) High Performance Extreme Computing Conference (HPEC), 2016 IEEE. :1-6 Sep, 2016
Subject
Communication, Networking and Broadcast Technologies
Computing and Processing
Computer architecture
Metadata
Distributed databases
Middleware
Collaboration
Computational modeling
distributed computing
risk
analytics
distributed data
open system architectures
data intensive computing
big data and distributed computing
Language
Abstract
This paper describes a multi-institution effort to develop a “data science as a service” platform. This platform integrates advanced federated data management for small to large datasets, access to high performance computing, distributed computing and advanced networking. The goal is to develop a platform that is flexible and extensible while still supporting domain research and avoiding the walled garden problem. Some preliminary lessons learned and next steps will also be outlined.