U.S. Department of Energy to Provide $6 Million for Research on Advanced Networking
Berkeley Lab’s Mariam Kiran to lead one of those projects
Contact: [email protected]
This week, the U.S. Department of Energy (DOE) announced a plan to provide $6 million in funding for five projects in basic research to advance 5G wireless networking for science applications. One of those projects is the “Large-scale, Self-driving 5G Network for Science,” led by Mariam Kiran, a research scientist at the Energy Sciences Network (ESnet) and Berkeley Lab’s Scientific Networking Division.
The nation’s emerging fifth-generation (5G) network is significantly faster than previous network generations and has the potential to improve connectivity across scientific infrastructure. Potential applications include linking remote experimental facilities and distributed sensing instrumentation with supercomputing resources to facilitate transporting and managing the huge volume of data generated by today’s scientific experiments.
Kiran’s project will use artificial intelligence (AI) combined with network virtualization to support complex end-to-end network connectivity — from edge 5G sensors to supercomputing facilities like the National Energy Research Scientific Computing Center (NERSC). These connections will be facilitated by ESnet.
“Our goal is to research and develop a flexible intelligent middleware that will allow all DOE science applications that are using edge wireless sensors to connect to DOE supercomputers for real-time or bulk data analysis,” said Kiran. “Where traditionally our research in ESnet has been focused on optic networks, we are advancing our methods and network tools to tackle wireless networks also, as we see these becoming more and more important for future DOE science. We aim to make it easy for scientists using 5G technologies to have access to an advanced network backend advancing a new digital continuum.”
Kiran noted that this project will allow DOE research projects that are using drones, mobile technologies, temperature sensors, or any wireless communication to connect seamlessly to ESnet.
“Along with the networking challenges, we are also exploring the trust and security of the data coming from the sensors and how AI can help manage 'bad' data actors to build close relationships with reliable baselines and 5G edge-to-core connections,” she added.
“Telecommunication networks based on 5G technologies have the potential to transform how we design, build, operate, and optimize scientific facilities and experiments,” said Barbara Helland, Associate Director of Science for Advanced Scientific Computing Research (ASCR). “Advanced wireless networks will make scientific facilities more mobile and agile while creating a pathway to the development of new sensing instrumentation for the collection of data in remote, inaccessible locations.”
"Advanced wireless networks like 5G combined with high-speed scientific optical networks like ESnet offer the promise of solving the 'last mile problem' for field science, creating new ways for researchers to connect data from sensors, instruments, and experiments in isolated locations with DOE's world-class supercomputers," said Inder Monga, Director of ESnet and Berkeley Lab's Scientific Networking Division. "At ESnet, our mission has always been to provide cutting-edge network technologies to accelerate scientific breakthroughs. This funding will allow us to explore how to leverage advanced wireless for future science."
Projects were chosen by competitive peer review under DOE Funding Opportunity Announcement “5G Enabled Energy Innovation Advanced Wireless Networks for Science,” sponsored by ASCR.
A list of awards can be found on the ASCR homepage under the heading “What’s New.”
Read the DOE press release here: https://www.energy.gov/science/articles/us-department-energy-announces-6-million-towards-5g-advanced-wireless-networks
More on Kiran’s work: https://www.es.net/news-and-publications/esnet-news/2020/machine-learning-to-add-another-dimension-to-esnets-toolbox-for-predicting-data-patterns/