Photo of Qin Zhang
Qin Zhang, an assistant professor at the School of Informatics, Computing, and Engineering, has been awarded a CAREER grant from the National Science Foundation to develop communication-efficient solutions for distributed computation and monitoring that will lower the cost of communication of data.
Zhang’s proposal, “Foundation of Communication-Efficient Distributed Computation and Monitoring,” will study fundamental algorithmic problems in databases, data mining, networking, and machine learning with the aim of creating a systematic theory of communication-efficient computing and monitoring that can have the potential to impact a wide range of rapidly developing areas in the theoretical foundations of big data.
“The CAREER award is one of the NSF’s most prestigious awards in support of early-career faculty, and I am certainly very excited about it,” Zhang said. “I have worked in this area for about 10 years, and I was aware that many fundamental questions in this area are still open. I’m happy that the design of communication-efficient algorithms is being recognized as an important direction for the future, and my work will hopefully generate a big impact in this area.”
This project targets three fundamental aspects of communication-efficient distributed computation, including the tradeoffs between the communication cost and the number of rounds of the computation in distributed one-shot computation, the power of data partitioning, and the connections between distributed one-shot computation and continuous monitoring. The results of the project will be integrated into a trilogy of courses in the foundations of data science to train students to tackle the communication bottlenecks of the future.
“This proposal is based on the hypothesis that as the scale of big data analytics continues to grow, communication will become a more critical resource than space and time,” Zhang said. “Communication-efficient solutions will be of central importance in the algorithm design. Basically, if the data is stored in different locations, then any computation on the global data needs communication. We will try to settle the communication complexity of fundamental problems in computer science, such as clustering, matrix multiplication, linear programming, and more.”
Zhang’s research will focus on the theoretical side of computing, but all algorithmic results developed in this project will have immediate impact on practical applications since the distributed one-shot computation and monitoring models are abstracted from numerous real-world applications.
“The problem of efficiently communicating the wealth of data that is created on a daily basis is one of the bigger challenges of our future,” said Kay Connelly, associate dean for research at SICE. “Qin’s work will help further our understanding of how we can attack this problem, and it is another sign of our leadership in the critical area of theoretical computer science.”