Center for Cloud Research and Development, NII

The National Institute of Informatics (NII) seeks to advance integrated research and development activities in information-related fields, including networking, software, and content. NII also promotes the creation of a state-of-the-art academic information infrastructure, such as SINET, a Japanese high-performance R&E network.

The center aims at establishment of the national advanced R&E infrastructure taking advantage of clouds. We research and develop advanced cloud infrastructure technologies such as inter-cloud, which utilizes multiple cloud platforms in a federated manner.

This website presents the latest results of our projects aimed at building the academic inter-cloud.

SINET5 NSI

Projects

This project develops new infrastructure technology, application-centric overlay cloud, to automatically build data analysis platforms optimized for and dedicated to user applications. Our goal also includes building an inter-cloud testbed configured by production-ready computing/networking resources and running applications of genome sequencing (distributed data processing) and fluid acoustic analysis (coupled simulation). We develop middleware to automatically build data analysis platforms (Virtual Cloud Provider) and a mechanism for dynamically configuring computing resources to optimize resource allocation for an application.

Big data generated by many indoor and outdoor IoT devices need to be securely analyzed in real-time for creation of various innovative services. SINET allows construction of an end-to-end isolated HPC and IoT environment using VPN over mobile and wired networks. We develop SINETStream, a software library enabling easy development of secure and efficient IoT applications over the environment.

It is as important to share infrastructure design and detailed reproducible workflows with participants as to actually automate complex operations. Literate computing is an approach utilizing Jupyter Notebook both to describe automated operations as live code and to share reproducible outcomes as technical and non-technical information in a manner similar to narrative stories. We develop reproducibility extensions for IT administrators.

The adoption of cold storage services should be considered in order to reduce the TCO and the labor of storage management of maintaining large amounts of data for long periods. The goal of this project is to acquire practical information for making decisions about storing research data in cloud cold storage and designing an overall data storage architecture by conducting experiments using the cold storage services of multiple commercial public clouds.