Return to Directory

Sushil Prasad


B.Tech. (Honors), Computer Science and Engineering, Indian Institute of Technology Kharagpur, 1985
M.S., Computer Science, Washington State University, 1986
Ph.D., Computer Science, University of Central Florida, 1990


Parallel and distributed computing and systems: parallel and distributed algorithms and data structures, distributed algorithms for sensor networks, parallel discrete event simulation, middleware and collaborative applications for heterogeneous mobile devices, web-based distributed and collaborative computing and workflows


Within my research area of parallel and distributed computing, my primary subareas of focus are parallel data structures and algorithms, parallel discrete event simulation, and nonnumeric computations including those on load balancing, graph algorithms, and parentheses matching. I have chosen to carry out fundamental research, along with practical implementations, for some key problems, which most often have been outstanding for a decade or more. Perhaps my best-known body of work leading to key advances in parallel data structures and discrete event simulation is the first theoretically scalable and currently the best practical event queue data structure for shared-memory architectures, namely, “Parallel Heaps.” This is accompanied by several efficient shared-memory algorithms for optimistic and conservative simulation which have resulted in parallel software systems with effective speedups for hard-to-parallelize simulations, such as those for VLSI logic circuits. My optimistic simulation algorithms have been refined to a stage now that the long outstanding problems of frequent rollbacks and/or of unbounded amount of storage for check-pointing that have plagued all previous algorithms in the last decade have been drastically reduced to almost no rollbacks and just one checkpoint per entity.

I have also taken up emerging problems and broken significant new ground. These include my recent middleware work, namely System on Mobile Devices (SyD), for collaborative distributed computing over networked, heterogeneous, and possibly mobile devices and data sources, including Web services. SyD’s web coordination bond artifacts not only solve the outstanding problems of creating travel and meeting schedules with automatic triggering, renegotiations, and rescheduling, but have also been shown to be capable of modeling Petri nets and expressing all the established workflow and communication patterns. The SyD middleware and the resulting series of work has resulted in two Ph.D. dissertations and several M.S. theses. Recently, some of the problems we are exploring include real-time collaborative editing with several recent publications, adaptive Quality-of-Service architectures, with flexible quality-of-security and transactions over heterogeneous data sources, identification of a set of core artifacts needed for Web service coordination and object composition for rapid modeling and execution of group transaction, work flows, and virtual organizations, with applications to domains ranging from e-commerce, to homeland security to biomedical experiments, lifetime problem of sensor networks, and platform technologies for high-performance parallel and distributed applications.

In addition to the problems highlighted earlier, some of the problems we are currently exploring include (i) exploratory frameworks for distributed algorithms for optimization problems on networks of heterogeneous sensors—an NSF EAGER grant has just been received for next two years to pursue this; (ii) multi-core and GPU-based data structures and algorithms for priority-queue based applications—collaboration with two Oak Ridge National Lab (ORNL) groups has been initiated; (iii) scalable computation on distributed geospatial data—a collaborative NSF proposal with Georgia Tech and the University of Illinois, Urbana-Champaign, has been submitted; (iv) P2P unstructured search algorithms—a dissertation is ongoing; and (v) porting/translating message-passing interface (MPI) programs to GPUs—another dissertation is ongoing on this, and collaboration with a third group at ORNL on GPU-based molecular modeling has been initiated.

I am also currently involved with biocomputing (modeling and simulation of proteins) and neuroinformatics (a knowledge base for neurons and their integration, funded by NIH and B&B) collaborative projects, related to my associate faculty position at Georgia State’s Neuroscience Institute.


X. Wang, F. Qiu, S. K. Prasad, and G. Chen, Efficient parallel algorithms for maximum-density subsequence problem, Proceedings of the 24th IEEE International Parallel and Distributed Processing Symposium (IPDPS 2010), Atlanta, April 2010.

A. Dhawan and S. K. Prasad, Taming the exponential state space of the maximum lifetime sensor cover problem, Proceedings of the 16th International Conference on High Performance Computing (HiPC 2009), Cochin, India, December 2009.

A. Sulistio, U. Cibej, S. K. Prasad, and Rajkumar Buyya, GarQ: an efficient scheduling data structure for advance reservations of grid resources, International Journal of Parallel, Emergent and Distributed Systems (IJPEDS), vol. 24, no. 1, 2009, pp. 1–19.

J. A. Preston and S. K. Prasad, P2P document tree management in a real-time collaborative editing system, Proceedings of the 14th International Conference on High Performance Computing (HiPC), Goa, India, December 2007, Lecture Notes in Computer Science, vol. 4873, pp. 418–431.

H. Botadra, Q. Cheng, S. K. Prasad, E. Aubanel, and V. Bhavsar, iC2mpi: a platform for parallel execution of graph-structured iterative computations, 21st IEEE International Parallel and Distributed Processing Symposium (IPDPS 2007) Workshops, Long Beach, March 2007.

B. Liu, S. K. Prasad, and E. Dogdu, A small listener for heterogeneous mobile devices: a service enabler with a uniform Web object view, Proceedings of the 2005 International Conference on Web Services (ICWS 2005), Orlando, July 2005.

S. K. Prasad and J. Balasoorya, Fundamental capabilities of Web coordination bonds: modeling Petri nets and expressing workflow and communication patterns over Web services, Hawaii International Conference on System Sciences (HICSS-38), Big Island, Hawaii, January 2005.

S. K. Prasad, V. Madisetti, S. Navathe, et al., System on Mobile Devices (SyD): a middleware testbed for collaborative applications over small heterogeneous devices and data stores, Proceedings of the ACM/IFIP/USENIX 5th International Middleware Conference (Middleware 2004), Toronto, October 2004, pp. 352–371.

S. K. Prasad and Z. Cao, Syncsim: a synchronous simple optimistic simulation technique based on a global parallel heap event queue, Proceedings of the 2003 Winter Simulation Conference (WSC 2003), New Orleans, December 2003.

S. K. Prasad and N. Junankar, Parallelizing a sequential logic simulator using an optimistic framework based on a global parallel heap event queue: an experience and performance report, Proceedings of the 14th Workshop on Parallel and Distributed Simulation (PADS 2000), Bologna, Italy, May 2000.

N. Deo, M. Medidi, and S. K. Prasad, Load balancing in parallel battlefield simulation on local- and shared-memory architectures, International Journal of  Computer Systems Science & Engineering, vol. 13, no. 1, 1998, pp. 55–65.

S. K. Prasad and S. Sawant, Parallel heap: a practical priority queue for fine-grained applications on small multiprocessors, Proceedings of the 7th IEEE Symposium on Parallel and Distributed Processing (SPDP ’95), San Antonio, October 1995, pp. 328–335.

S. K. Prasad, S. K. Das, and C.-Y. Chen, Efficient EREW PRAM algorithms for matching parentheses, IEEE Transactions on Parallel and Distributed Systems, vol. 5, no. 9, 1994, pp. 995–1008.

S. K. Prasad, Efficient and scalable PRAM algorithms for discrete event simulation of bounded degree networks, Journal of Parallel and Distributed Computing, vol. 18, no. 4, August 1993, pp. 524–530.

S. K. Das, N. Deo, and S. K. Prasad, Two minimum spanning forest algorithms for fixed-size hypercube computers, Parallel Computing, vol. 15, 1990, pp. 179–187.

N. Deo and S. K. Prasad, Parallel heap: an optimal parallel priority queue, Journal of Supercomputing, vol. 6, 1992, pp. 87–98.