Tevfik Kosar Program Director, NSF/CISE/OAC Professor of Computer Science & Engineering Director, Data Intensive Distributed Computing Laboratory Department of Computer Science & Engineering University at Buffalo (SUNY) 338-J Davis Hall Buffalo, NY 14260-2000 Phone: 716-645-2323 Email: tkosar (at) buffalo.edu twitter.com/TevfikKosar linkedin.com/TevfikKosar |
![]() |
Biography
My research focuses on applying advanced AI/ML techniques to optimize the performance, scalability, and efficiency of distributed computing systems and big-data analytics pipelines. My significant contributions are in the areas of geo-distributed data analytics, performance and energy efficiency in large-scale distributed systems, wide-area distributed coordination, network and I/O optimization, and end-to-end dataflow management. I designed and developed the Stork distributed data scheduling system which is used by institutions worldwide; and the PetaShare distributed storage network that managed close to a Petabyte of storage located across nine university campuses in Louisiana. I am currently working on data scheduling and optimization as a cloud-hosted service (OneDataShare); and energy-efficient data transfers for green Internet (GreenDataFlow). Major application areas of my research include genomics, health sciences, emergency response, and coastal and environmental sciences.
Selected Awards:
- UB Exceptional Scholar: Sustained Achievement Award, 2020
- UB CSE Outstanding Service Award, 2019
- IEEE Region-I Technological Innovation Award, 2019
- UB SEAS Senior Researcher of the Year, 2018
- UB SEAS Senior Teacher of the Year, 2018
- IBM Research Award, 2017
- UB CSE Senior Faculty Research Award, 2017
- UB CSE Senior Faculty Teaching Award, 2016
- Best Paper Award, ICECCO 2013
- Best Paper Award, IEEE/ACM SC'12 NDM Workshop, 2012
- Google Research Award, 2011
- NSF CAREER Award, 2009
- Bussiness Report's Top 40 Under 40, 2009
- LSU Rainmaker Award, 2009
- LSU Flagship Faculty, 2009
- 1012 Corridor Young Scientist, 2009
- LSU College of Basic Sciences Research Award, 2009
- LSU CCT Faculty of the Year Award, 2008
Selected Publications:
-
Z. Nine and T. Kosar.
A Two-Phase Dynamic Throughput Optimization Model for Big Data Transfers.
In IEEE Transactions on Parallel and Distributed Systems (TPDS), DOI:10.1109/TPDS.2020.3012929 (2020).
-
A. Ailijiang, A. Charapko, M. Demirbas, and T. Kosar.
WPaxos: Wide Area Network Flexible Consensus.
In IEEE Transactions on Parallel and Distributed Systems (TPDS), Vol.31 No.1 (2020), pp.211-223.
-
E. Arslan and T. Kosar.
High-Speed Transfer Optimization Based on Historical Analysis and Real-Time Tuning.
In IEEE Transactions on Parallel and Distributed Systems (TPDS), Vol.29 No.6 (2018), pp.1303-1316.
-
E. Yildirim, E. Arslan, J. Kim and T. Kosar. Application-Level Optimization of Big Data
Transfers Through Pipelining, Parallelism and
Concurrency.
In IEEE Transactions on Cloud Computing (TCC), Vol.4 No.1 (2016) pp. 63-75.
-
E. Arslan, K. Guner, and T. Kosar.
HARP: Predictive Transfer Optimization Based on Historical Analysis and Real-time Probing.
In Proceedings of IEEE/ACM Supercomputing (SC), 2016.
-
I. Alan, E. Arslan, and T. Kosar.
Energy-Aware Data Transfer Algorithms.
In Proceedings of IEEE/ACM Supercomputing (SC), 2015.
-
D. Yin, E. Yildirim, S. Kulasekaran, B. Ross, and T. Kosar. A Data Throughput Prediction and Optimization Service for Widely Distributed Many-Task Computing.
In IEEE Transactions on Parallel and Distributed Systems (TPDS), Vol.22 No.6 (2011) pp.899-909.
-
E. Yildirim, D. Yin, and T. Kosar. Prediction of Optimal Parallelism Level in Wide Area Data Transfers.
In IEEE Transactions on Parallel and Distributed Systems (TPDS), Vol.22 No.12 (2011) pp.2033-2045.
-
T. Kosar, M. Balman, E. Yildirim, S. Kulasekaran and B. Ross.
Stork Data Scheduler: Mitigating the Data Bottleneck in e-Science.
In Philosophical Transactions of the Royal Society A, Vol.369 (2011), pp. 3254-3267.
-
T. Kosar, I. Akturk, M. Balman, and X. Wang.
PetaShare: A Reliable, Efficient, and Transparent Distributed Storage Management System.
In Scientific Programming Journal (SPJ), Vol. 19 No.1 (2011), pp.27-43.
A full list of publications can be found here.