Bridging the gap between theory and practice for distributed graph algorithms

2023 Fiscal Year Final Research Report

• Project/Area Number: 21KK0204
• Research Category: Fund for the Promotion of Joint International Research (Fostering Joint International Research (A))
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 60010:Theory of informatics-related
• Research Institution: Japan Advanced Institute of Science and Technology
• Principal Investigator: Schwartzman Gregory 北陸先端科学技術大学院大学, 先端科学技術研究科, 准教授 (20815261)
• Project Period (FY): 2022 – 2023
• Keywords: distributed computing / networking / graph algorithms

Outline of Final Research Achievements

During the duration of this research project we managed to used theoretical tools from the field of distributed graph algorithms to speed up primitives used in networking. Specifically, we show how to use theoretical tools such as graph spanners to create networking algorithms with a fast recovery time from failures. Our results apply to fundamental problems such as multicast and clock synchronization.

We also introduced a new theoretical model for distributed graph algorithms that limits the amount of space used by the algorithms. This is critical to allow for a smoothed transition of theory algorithms to practice, as space is a major bottleneck in data centers. We currently have 2 papers in submission.

Free Research Field

distributed computing

Academic Significance and Societal Importance of the Research Achievements

Data centers are a critical infrastructure in today's society. They are used by all major companies such as Google, Amazon, etc...
Speeding up the algorithms used in these data centers can lead to a huge financial and societal impact (e.g., faster cloud computing).