The Aida Laboratory is a research group devoted to the study of parallel and distributed computing systems.  Our goal is to develop fundamental/practical technologies to effectively utilize IT resources in parallel and distributed computing platforms, including clusters, grids and clouds. Currently, we focus on the following projects:

Distributed computing infrastructure

This project studies architecture of distributed computing infrastructure. We develop software to build and operate the infrastructure, including testbeds for R&D and infrastructure for the production-ready operation. Our research themes include: an authentication system and middleware for distributed computing infrastructure. 

      recent publications/talks:

  • Shigetoshi Yokohama, Yoshinobu Masatani, Tazro Ohta, Osamu Ogasawara, Nobukazu Yoshioka, Kai Liu, Kento Aida, Reproducible Scientific Computing Environment with Overlay Cloud Architecture, Proceedings of the 9th IEEE International Conference on Cloud Computing (IEEE CLOUD 2016), Jul. 2016
  • Kai Liu, Kento Aida, Shigetoshi Yokoyama, Yoshinobu Masatani, Flexible Container-Based Computing Platform on Cloud for Scientific Workflows, Proceedings of the International Conference on Cloud Computing Research and Innovation (ICCCRI2016), May 2016
  • Shigeo Urushidani, Shunji Abe, Kenjiro Yamanaka, Kento Aida, Shigetoshi Yakoyama, Hiroshi Yamada, Motonori Nakamura, Kensuke Fukuda, Michihiro Koibuchi, Shigeki Yamada, New Directions for a Japanese Academic Backbone Network, IEICE Transactions on Information and Systems, E98-D(3), pp.546-556, Feb. 2015
  • Kento Aida, Advanced Academic Information Infrastructure Utilizing Clouds, The 6th Annual International Conference on Information and Communication Technology for Embedded Systems (IC-ICTES 2015), Mar. 2015 [slide]
  • Eisaku Sakane, Kento Aida, Kazutaka Motoyama, Design and Implementation of Certificate Authority for High Performance Computing Infrastructure in Japan, Proceedings of Science, The International Symposium on Grids and Clouds (ISGC 2013), Feb. 2014
  • NAREGI-CA [web]

Scheduling and resource management

This project studies novel scheduling algorithms for parallel and distributed computing systems. We also develop software for scheduling and resource management. Research themes include: parallel job scheduling algorithms on a cluster, interactive scheduling software on grids and clouds, market-based resource allocation algorithms on clouds, load balancing algorithms for inter-clouds, and scheduling for MapReduce applications.

      recent publications:

  • Hayata Ohnaga, Kento Aida, Omar Abdul-Rahman, Performance of Hadoop Application on Hybrid Cloud, Proceedings of the International Conference on Cloud Computing Research and Innovation (ICCCRI2015), Oct. 2015
  • Kento Aida, Omar Abdul-Rahman, Eisaku Sakane, Kazutaka Motoyama, Long-term Performance Evaluation of Hadoop Jobs in Public and Community Clouds, IEICE Transactions on Information and Systems, E98-D(6), pp.1176-1184, Jun. 2015
  • Omar Abdul-Rahman, Kento Aida, Towards Understating the Behavior of Users in Utilizing Google Cloud: Mice and Elephants Phenomena, Proceedings of the 6th IEEE International Conference on Cloud Computing Technology and  Science (CloudCom 2014), pp.272-277, Dec. 2014

Application software

This project develops application software that runs efficiently on parallel and distributed computing systems. Research themes include: optimization using parallel branch and bound algorithms, optimization of baseball strategy, crop parameter identification using satellite images, and astrophysical simulation.

      recent publications:

  • Kazutaka Motoyama, Yoshikazu Tanaka, Kento Aida, Eisaku Sakane, Kenichi Miura, Effective system for simulating dust continuum observations on distributed computing resources, Journal of Space Science Informatics Japan, 3(), pp.155-161, Mar. 2014