Download PDF
DataDirect Networks > Case Studies > Changing Research with a Forward-Looking AI and Big Data Computing Infrastructure
DataDirect Networks Logo

Changing Research with a Forward-Looking AI and Big Data Computing Infrastructure

Technology Category
  • Analytics & Modeling - Big Data Analytics
  • Analytics & Modeling - Machine Learning
  • Infrastructure as a Service (IaaS) - Cloud Computing
Applicable Industries
  • Education
Applicable Functions
  • Product Research & Development
Use Cases
  • Machine Condition Monitoring
  • Predictive Maintenance
Services
  • Cloud Planning, Design & Implementation Services
  • Data Science Services
The Challenge
Tokyo Institute of Technology (Tokyo Tech) was faced with the challenge of speeding up data access times in parallel with continually improving algorithms that interact with data subsystems. They aimed to achieve this while maintaining optimal power consumption and system efficiency. The institution sought to break away from the conventions of the world's top supercomputers by incorporating elements and design points from containerization, cloud, artificial intelligence (AI), and Big Data.
About The Customer
Tokyo Institute of Technology (Tokyo Tech) is the largest institution for higher education in Japan dedicated to science and technology. The institution is known for its innovative approach to research and education in the fields of science and technology. Tokyo Tech is committed to advancing research computing for data-centric infrastructures for the future of research big data but also AI and machine learning. The institution's new TSUBAME3.0 system is a testament to this commitment, showcasing extreme innovation in the area of power consumption and system efficiency.
The Solution
To address the challenge, Tokyo Tech implemented a 15.9PB Lustre* parallel file system, composed of three DDN ES14K storage appliances. This solution was integrated into their new TSUBAME3.0 system, which breaks with many of the conventions of the world's top supercomputers. The system incorporates elements and design points from containerization, cloud, artificial intelligence (AI), and Big Data. The solution was designed to speed up data access times in parallel with continually improving algorithms that interact with data subsystems, all while achieving optimal power consumption and system efficiency.
Operational Impact
  • The TSUBAME3.0 system, with its integrated 15.9PB Lustre* parallel file system, is rated at a peak performance of 150GB/s.
  • The system has established a clear vision of advancing research computing for data-centric infrastructures for the future of research big data but also AI and machine learning.
  • The implementation of the system creates opportunities to transition to new application areas, such as graph computing and machine learning.
Quantitative Benefit
  • Peak performance of the system is 150GB/s.

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.