activity
20242026
collaborators

6 papers

cs.PL2026

Splyce: SIMD Vectorization of Sparse Coiteration

Kabilan Mahathevan, Poorna Gunathilaka, Kirshanthan Sundararajah

Sparse tensor contractions are bottlenecked by sparse-sparse coiteration loops that resist standard loop vectorization. We present Splyce, an auto-vectorization framework in MLIR t…

cs.DC2026

A Preliminary Study on Simultaneous Coscheduling for Discrete GPU vs. Fused GPU

Poorna Gunathilaka, Nabayan Chaudhury, Kirshanthan Sundararajah +1

CPU-GPU coscheduling enables simultaneous execution of an application across both processing units, but its efficiency depends on workload partitioning and memory architecture. Thi…

cs.PL2026

nomp: A Framework for Building Domain Specific Compilers

Thilina Ratnayaka, Kaushik Kulkarni, Nipuna Fernando +8

The low-level GPU programming models (CUDA, HIP, OpenCL, etc.) provide detailed control of the data flow and execution plan of a program in order to extract close-to-metal performa…

cs.SE2026

Assessing Large Language Models for Stabilizing Numerical Expressions in Scientific Software

Tien Nguyen, Kirshanthan Sundararajah, Muhammad Ali Gulzar

Scientific software relies on high-precision computation, yet finite floating-point representations introduce precision errors that propagate in safety-critical domains. Despite gr…

cs.PL2025

Eliminate Branches by Melding IR Instructions

Yuze Li, Srinivasan Ramachandra Sharma, Charitha Saumya +2

Branch mispredictions cause catastrophic performance penalties in modern processors, leading to performance loss. While hardware predictors and profile-guided techniques exist, dat…

cs.DC2024

Bring Your Own Formats and Kernels: Composable Abstractions for Sparse Matrix Computation

Pratyush Das, Amirhossein Basareh, Artem Pelenitsyn +3

Real-world sparse matrices often feature multiple forms of structured sparsity -- rectangular dense blocks, diagonal bands, and scattered entries -- that no single storage format c…