Systems & Research

Research Projects

Selected public research and software infrastructure. Projects still under anonymous review are not detailed here.

HPC-Bench
NeurIPS 2026 E&D · Accepted

A benchmark of 114 HPC workloads across 16 computational motifs evaluating LLM-driven optimization in serial CPU, OpenMP and CUDA. It combines reproducible builds, correctness checks and the Fast@k / Speedup@k metrics. The study finds that reliable optimization becomes harder as required speedup increases, particularly on CUDA.

LLMsHPCBenchmarkingCUDA
TileBench
AACL 2026 · Accepted

A controlled benchmark comparing 45 operators in Triton and cuTile on NVIDIA B200. Matched semantics, autotuning, NCU/PTX diagnosis and LLM-generated kernels expose workload-dependent compiler and hardware performance differences.

TritoncuTileGPU ProfilingB200
HPC-Performance-AI
DOE Genesis Mission · Ongoing

An ongoing framework for performance prediction and optimization in HPC applications. The open benchmark infrastructure spans GPU microbenchmarks, proxy/mini-applications and production scientific codes, with standardized build, execution, timing and validation workflows. Performance prediction and optimization components remain active research goals.

HPCGPUReproducibilityScientific Applications
LLM-Based HPC Optimization Study
ICPP Companion 2025

An earlier study of 24 HPC programs across nine computational domains, comparing LLM-driven serial and parallel optimization against compiler-based tools. Its correctness and performance findings motivated the expanded HPC-Bench evaluation.

LLMsCode OptimizationHPC