The principle studies of this lab are in **High Performance Computing** with applications to
Data Science and Parallel Discrete-Event Simulation (PDES). The work in data science focuses on
two areas, namely: (i) *Topological Data Analysis* and (ii) *Privacy Preserving Data
Mining*. The work focuses on the construction of high performance computing solutions to
each of these problems. While much of the focus of our work is with parallel and distributed
computing, we also explore randomized and approximate computing methods to accelerate the
problems in question.
The work in PDES focuses on *Optimistically (Time Warp) Synchronized PDES*. Our work is
primarily with the Time Warp Mechanism. Optimistically synchronized simulators do not strictly
enforce the causality relations between events during event processing. Under the Time Warp
mechanism, an optimistic simulation solution aggressively processes events and incorporates a
rollback recovery mechanism to use whenever a causal violation is discovered. Our studies
address parallel simulation on multi-core/many-core nodes and clusters.

- Topological Data Analysis and the Computation of Persistent Homology
- Scalable Big Data Clustering by Random Projection Hashing

The following document may help you as you work through the design and optimization of parallel and distributed systems: Butler Lampson, "Hints and Principles for Computer System Design, August 13, 2020.

Below are examples of previous projects that my students and I have worked on. While I have numerous other previous projects, these ideas remaining interesting to me but, for one reason or another, I am no longer listing them as active projects in my research program.