(PP20): An Implementation of International Data Cloud
AI/Machine Learning/Deep Learning
Clouds and Distributed Computing
TimeTuesday, June 18th3:15pm - 3:45pm
DescriptionData needs to be shared and integrated to create values. The global data sharing between international partners requires a sustainable trust-based platform. In 2018, an international data cloud effort between AIST Japan and NCHC Taiwan has been initiated to prototype a platform for sharing both the data and the computing power.
Since the Data Cloud has the nature of distributed data across organizational boundaries, it has to tackle with data sharing and safety, user management, etc.
There are 4 major tasks to be tackled with, they are 1). Connecting both data centers, 2). Secured data sharing and exchange, 3). Accessing data center from AIoT devices, 4). User authentication and authorization.
This 4-year project was set out to tackle the issues above and provide a prototype of data cloud service with AI facilities integrated. In addition, an AI application is identified to demonstrate the feasibility of such a data cloud implementation