About
Pramod Ganapathi is a Research Assistant Professor in the Department of Computer Science at Stony Brook University. His research interests include mathematical and algorithmic puzzles, algorithms to discover algorithms, algorithm design techniques, parallel algorithms, cache-efficient algorithms, big data, and high-performance computing. He has a strong background in mathematics, algorithms, and puzzles, with extensive experience in learning, teaching, and researching these areas. Ganapathi has published a book and over 15 papers in reputed international conferences and journals, has been granted a US patent, and has received two Outstanding Paper Awards. He has taught more than 10 courses across 25 offerings and supervised over 90 master's and bachelor students. Prior to his current position, he was an Assistant Professor at the Indian Institute of Technology, Indore, and before that, he founded and served as CEO of an animation-based online higher education startup called 'Learning is Beautiful' in India. He completed his Ph.D. in computer science at Stony Brook University, specializing in parallel algorithms, with a dissertation titled 'Automatic Discovery of Efficient Divide-and-Conquer Algorithms for Dynamic Programming Problems,' supervised by Prof. Rezaul A. Chowdhury. His research includes developing frameworks such as Autogen, Autogen-Wave, and Autogen-Fractile, which focus on automatically generating parallel divide-and-conquer dynamic programming algorithms that are energy-efficient, cache-oblivious, processor-oblivious, and architecture-independent. Currently, he is involved in a project centered on a mathematical and algorithmic puzzles book, which presents counterintuitive puzzles and solutions designed to challenge and educate through a puzzle-centric approach.
Research topics
- Computer Science
- Artificial Intelligence
- Parallel computing
- Algorithm
- Mathematics
- Geometry
- Computational science
- Database
Selected publications
Speeding up Stencil Computation using Gaussian Approximations
Society for Industrial and Applied Mathematics eBooks · 2025-01-01
book-chapterStencils are widely used in scientific and industrial computing for the simulation of physical systems. Given a multidimensional spatial grid containing initial data, these stencil patterns are applied uniformly to all cells of the grid over multiple timesteps to obtain the final data.
Speeding up Stencil Computation using Gaussian Approximations
Society for Industrial and Applied Mathematics eBooks · 2025-01-01 · 1 citations
book-chapterStencils are widely used in scientific and industrial computing for the simulation of physical systems. Given a multidimensional spatial grid containing initial data, these stencil patterns are applied uniformly to all cells of the grid over multiple timesteps to obtain the final data.
SSRN Electronic Journal · 2024-01-01
articleOpen accessUTTAR PRADESH JOURNAL OF ZOOLOGY · 2024-07-16
articleOpen accessThe flies of the genus Haematobia (Diptera:Muscidae) are hematophagous ectoparasites of medical and veterinary importance. The morphological identification of these flies is often complicated due to their similarities. This was the first documentation of Haematobia irritans exigua in Tamil Nadu which affected free-ranging Bargur Cattle (Bos indicus) from Bargur Hills of Tamil Nadu. The aim of this research was to provide the details on the molecular characterization and molecular divergence of the cytochrome c oxidase subunit 1 gene (COI) and 28S Ribosomal RNA of Haematobia irritans exigua and to compare them with the morphological features. The results of this study showed that the nucleotide composition of the COI gene of Haematobia irritans exigua had a higher AT (69.5%) and a lower GC (30.4%). The rate of transition was higher than that of transversion. The intra-specific distance based on COI gene analysis did not exceed 1.1%, while the inter-specific distance between different species ranged from 0.9% to 14.1%. A neighbor-joining tree and an Unweighted Pair Group Method with Arithmetic Mean (UPGMA) tree were constructed using 1000 bootstrapped samples to analyze the phylogenetic relationship between different muscidae species based on the COI gene as an identification marker. The analysis of 28S Ribosomal RNA showed an intra-specific distance of 1.6% and an inter-specific distance ranging from 1.49% to 8.56%. The 28S rRNA revealed the presence of intra-specific variation among the species. Additionally, this study found notable color variations on the thorax of the flies in the same population, along with a minute phenotypical variation with minimum intraspecific distance among the flies in the same population. This taxonomic data of the hematophagous fly Haematobia irritans exigua would be well-intentioned for identification.
A Fast Algorithm for Aperiodic Linear Stencil Computation using Fast Fourier Transforms
ACM Transactions on Parallel Computing · 2023-07-24 · 2 citations
articleStencil computations are widely used to simulate the change of state of physical systems across a multidimensional grid over multiple timesteps. The state-of-the-art techniques in this area fall into three groups: cache-aware tiled looping algorithms, cache-oblivious divide-and-conquer trapezoidal algorithms, and Krylov subspace methods. In this article, we present two efficient parallel algorithms for performing linear stencil computations. Current direct solvers in this domain are computationally inefficient, and Krylov methods require manual labor and mathematical training. We solve these problems for linear stencils by using discrete Fourier transforms preconditioning on a Krylov method to achieve a direct solver that is both fast and general. Indeed, while all currently available algorithms for solving general linear stencils perform Θ( NT ) work, where N is the size of the spatial grid and T is the number of timesteps, our algorithms perform o ( NT ) work. To the best of our knowledge, we give the first algorithms that use fast Fourier transforms to compute final grid data by evolving the initial data for many timesteps at once. Our algorithms handle both periodic and aperiodic boundary conditions and achieve polynomially better performance bounds (i.e., computational complexity and parallel runtime) than all other existing solutions. Initial experimental results show that implementations of our algorithms that evolve grids of roughly 10 7 cells for around 10 5 timesteps run orders of magnitude faster than state-of-the-art implementations for periodic stencil problems, and 1.3× to 8.5× faster for aperiodic stencil problems. Code Repository: https://github.com/TEAlab/FFTStencils
2022-07-10 · 4 citations
articleStencil computations are widely used to simulate the change of state of physical systems. The current best algorithm for performing aperiodic linear stencil computations on a d (≥ 1)-dimensional grid of size N for T timesteps does Θ(TN1-1/d+N Log N) work. We introduce novel techniques based on random walks and Gaussian approximations for an asymptotic improvement of this work bound for a class of linear stencils. We also improve the span (i.e., parallel running time on an unbounded number of processors) asymptotically from the current state of the art.
FOURST: A code generator for FFT-based fast stencil computations
2022-05-01 · 4 citations
articleStencil computations are ubiquitous in modern grid-based physical simulations. In this paper, we present FOURST – a compiler to generate programs computing time iterated linear periodic and aperiodic stencil computations with fast Fourier transform methods. This paper outlines the design and implementation of the code generation approach in FOURST, to automatically generate FFT-based stencil solvers. We present experimental results on the state-of-the-art Ookami supercomputer housing Fujitsu A64FX and Intel Skylake processors, to study the performance of FOURST and a state-of-the-art tiling-based optimized code generator PLuTo on various stencil shapes and varying the number of time iterations. We discuss the performance profiles, and limitations, of both approaches on high-end modern hardware.
Parallel Divide-and-Conquer Algorithms for Bubble Sort, Selection Sort and Insertion Sort
The Computer Journal · 2021 · 7 citations
1st authorCorresponding- Computer Science
- Computer Science
- Parallel computing
Abstract We present efficient parallel recursive divide-and-conquer algorithms for bubble sort, selection sort, and insertion sort. Our algorithms have excellent data locality and are highly parallel. The computational complexity of our insertion sort is ${{\mathcal{O}}}\left ({n^{\log _2 3}}\right )$ in contrast to ${{\mathcal{O}}}\left ({n^2}\right )$ of standard insertion sort.
Fast Stencil Computations using Fast Fourier Transforms
2021 · 11 citations
- Computer Science
- Computer Science
- Parallel computing
Stencil computations are widely used to simulate the change of state of physical systems across a multidimensional grid over multiple timesteps. The state-of-the-art techniques in this area fall into three groups: cache-aware tiled looping algorithms, cache-oblivious divide-and-conquer trapezoidal algorithms, and Krylov subspace methods.
Information Processing Letters · 2021-07-21 · 5 citations
articleSenior authorCorresponding
Frequent coauthors
- 32 shared
Rezaul Chowdhury
Stony Brook University
- 12 shared
Jesmin Jahan Tithi
- 11 shared
Yuan Tang
South China University of Technology
- 9 shared
Zafar Ahmad
Stony Brook University
- 8 shared
Rathish Das
- 7 shared
Aaron Gregory
Stony Brook University
- 5 shared
Mohammad Mahdi Javanmard
- 4 shared
Y. Zhu
Stony Brook University
Labs
Research interests include mathematical puzzles, algorithmic puzzles, algorithms, and algorithm design techniques.
Awards & honors
- Two Outstanding Paper Awards at SPAA 2021
- Undergraduate Education Award 2022
- Resume-aware match score
- Save to shortlist
- AI-drafted outreach
See your match with Pramod Ganapathi
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
- Free to start
- No credit card
- 30-second signup