Keqin Li - Selected Publications#

Dr. Keqin Li's extraordinary (897) accomplishments include 16 books/proceedings, 52 book chapters, 563 journal articles, 200 conference papers, and 66 patents. He has 17652 Google Scholar citations, with h-index = 64 and i10-index = 383. He is listed in the world's top 2% scientists by Elsevier's Mendeley Data and Scopus citation database in 2019-2021, using a composite indicator that considers and combines six citation metrics. Based on single-year (2020) impact, he is ranked #1 in the US and #3 in the world in parallel and distributed computing. Based on career-long impact (2021), he is ranked #3 in the US and #4 in the world.

Ten representative publications:

(1) B. Yi, X. Wang, K. Li, S. K. Das, and M. Huang, "A comprehensive survey of network function virtualization," Computer Networks, 133, 212-262, 2018. (IF=4.474, Goggle-Scholar-Citation=321)

This is a thorough, comprehensive, and influential survey of an important computer network research area, and immediately attracted extensive attention in the international research community. Dr. Li has contributed extensively to computer networking. His research has covered wireless communication networks, sensor networks, peer-to-peer file sharing systems, transport networks, software-defined networks, controller area networks, in-vehicle networks, networked storages, and Internet of things.

(2) J. Chen, K. Li, Z. Tang, K. Bilal, S. Yu, C. Weng, and K. Li, "A parallel random forest algorithm for big data in a spark cloud computing environment," IEEE Transactions on Parallel and Distributed Systems, 28 (4), 919-933, 2017. (IF=4.181, Goggle-Scholar-Citation=312)

This is a representative study integrating artificial intelligence, big data processing, deep learning, and parallel computing into cloud computing environments. Dr. Li has implemented numerous big data algorithms, machining learning approaches, and health care applications on various parallel and distributed systems such as in memory computing systems, heterogeneous CPU GPU clusters, and cloud computing environments. He created FlinkCL and GFlink to extend openCL and Flink on GPU accelerated heterogeneous distributed systems. He also developed HeteroYARN to extend YARN on FPGA accelerated heterogeneous architectures.

(3) M. Ali, R. Dhamotharan, E. Khan, S.U. Khan, A.V. Vasilakos, K. Li, and A.Y. Zomaya, "SeDaSC: Secure data sharing in clouds," IEEE Systems Journal, 11 (2), 395-404, 2017. (IF=5.28, Goggle-Scholar-Citation=188)

This was a unique study integrating data storage, data sharing, and data security in cloud computing. Dr. Li has been one of the most active and productive researchers in cloud computing and mobile edge computing. He has the most (21 so far) publications on IEEE-TCC, the flagship journal in the field. He studied could storage, cloud security, cloud reliability, service selection, and computation offloading, both theoretically and experimentally. He received the NPC-2019 and ISPA-2019 Best Paper Awards.

(4) Y. Xu, K. Li, J. Hu, and K. Li, "A genetic algorithm for task scheduling on heterogeneous computing systems using multiple priority queues," Information Sciences, 270, 255-287, 2014. (IF=6.795, Goggle-Scholar-Citation=341)

This is a representative study of task scheduling in heterogeneous systems, combining the techniques of genetic algorithms and heuristic algorithms. Dr. Li has made tremendous contributions to task scheduling for over 30 years. His early 1990 work of job scheduling on meshes has become classic. His long term and extensive research has covered task/job/workflow scheduling in a wide variety of parallel and distributed systems, including partitionable parallel systems, cluster and grid computing systems, multicore processors, heterogeneous computing systems, embedded systems, big data and cloud computing systems, cyber physical systems. He received the NAECON-1997 Best Paper Award (first place).

(5) K. Li, X. Tang, and K. Li, "Energy-efficient stochastic task scheduling on heterogeneous computing systems," IEEE Transactions on Parallel and Distributed Systems, 25 (11), 2867-2876, 2014. (IF=4.181, Goggle-Scholar-Citation=177)

This was a unique study involving energy efficiency, stochastic tasks, and heterogeneous computing systems. Dr. Li is an international leading expert in energy efficient computing for over 15 years. He adopted a unique research methodology, i.e., deriving lower bounds for optimal solutions and comparing solutions of heuristic algorithms with optimal solutions. He has conducted systematical investigation for sequential and parallel, independent and precedence constrained tasks, on single and multiple, homogeneous and heterogeneous processors with continuous and discrete, regular and irregular, bounded and unbounded speed levels. He won the ISPA-2016 Best Paper Award.

(6) J. Cao, K. Li, and I. Stojmenovic, "Optimal power allocation and load distribution for multiple heterogeneous multicore server processors across clouds and data centers," IEEE Transactions on Computers, 63 (1), 45-58, 2014. (IF=3.131, Goggle-Scholar-Citation=195)

This was one of the earliest papers in analytical study of power-performance tradeoff, and was highlighted by the Editor-in-Chief of IEEE-TC in Spotlight on Transactions of IEEE Computer, vol. 47, no. 5, pp. 6-7, May 2014. Dr. Li takes a unique approach to investigating various aspects of cloud computing, i.e., analytical modeling, quantitative analysis, and algorithmic and numerical optimization. He analytically formulated and algorithmically solved the problem of optimal power allocation and load distribution for multiple heterogeneous multicore servers across clouds and data centers, i.e., power constrained performance optimization and performance constrained power optimization. He investigated power-performance tradeoff and cost-performance ratio optimization on multiple heterogeneous servers with dynamic speed and power management.

(7) J. Cao, K. Hwang, K. Li, and A.Y. Zomaya, "Optimal multiserver configuration for profit maximization in cloud computing," IEEE Transactions on Parallel and Distributed Systems, 24 (6), 1087-1096, 2013. (IF=4.181, Goggle-Scholar-Citation=248)

This was the first paper in analytical study of profit maximization, and was the spotlight paper of the issue selected from over 100 submissions. Dr. Li rigorously established analytical models for profit maximization, extensively investigated data center power and performance tradeoff, and quantitatively characterized horizontal and vertical elasticity. He established analytical models (a multiserver model, power consumption models, and a profit model) for price, revenue, cost, and profit, so that profit maximization of a cloud service provider can be studied analytically and algorithmically. He introduced customer satisfaction, used a noncooperative game to formulate market competition and stabilization, and developed an algorithm to find the Nash equilibrium. He published the first comprehensive survey in this area.

(8) J. Li, H. Kameda, and K. Li, "Optimal dynamic mobility management for PCS networks," IEEE/ACM Transactions on Networking, 8 (3), 319-327, 2000. (IF=3.56, Goggle-Scholar-Citation=233)

This was one of the earliest papers on dynamic location management in wireless networks. Dr. Li has significantly and systematically contributed to dynamic mobility management in wireless communication networks for over 20 years. He deeply investigated the costs of location update and terminal paging in distance based, movement based, and time based location management schemes, and achieved so far the best theoretical and analytical results.

(9) Y. Pan and K. Li, "Linear array with a reconfigurable pipelined bus system—concepts and applications," Information Sciences 106 (3-4), 237-258, 1998. (IF=6.795, Goggle-Scholar-Citation=151)

This was the first paper introducing the LARPBS model. Dr. Li was an internationally recognized authority in parallel computing using optical interconnections. He proposed one of the first two models in this area and won a Best Paper Award from PDPTA-1996. He developed numerous fast and scalable parallel algorithms for a wide spectrum of applications and demonstrated their superior performance (i.e., speed and scalability). He was the leading and primary editor of the first book in the area of parallel architectures, algorithms, and applications with optical interconnections (1998).

(10) K. Li and K.H. Cheng, "A two dimensional buddy system for dynamic resource allocation in a partitionable mesh connected system," Proceedings of ACM Eighteenth Computer Science Conference, 22-27, 1990. (Goggle-Scholar-Citation=224)

This was the first paper initiating system partitioning and processor allocation. Dr. Li is the most internationally renowned researcher in processor allocation and job scheduling in mesh connected systems due to his seminal contributions in 1989-1991. He proposed a two dimensional binary buddy system (implemented on Intel Paragon) for dynamic processor allocation and reclamation. He developed and analyzed fast and efficient heuristic job scheduling algorithms. His pioneering work has initiated and inspired extensive subsequent studies by numerous investigators, and has created a very active and productive research field for 30+ years.