4 papers
Towards an Accurate GPU Data Race Detector
Ajay Nayak, Anubhab Ghosh, Arkaprava Basu
Data races in GPU programs pose a threat to the reliability of GPU-accelerated software stacks. Prior works proposed various dynamic (runtime) and static (compile-time) techniques…
Chasing Elusive Memory Bugs in GPU Programs
Anubhab Ghosh, Ajay Nayak, Dhananjay Rao Thallikar Shyam +1
Memory safety bugs, such as out-of-bound accesses (OOB) in GPU programs, can compromise the security and reliability of GPU-accelerated software. We report the existence of input-d…
PyGraph: Robust Compiler Support for CUDA Graphs in PyTorch
Abhishek Ghosh, Ajay Nayak, Ashish Panwar +1
Machine learning (ML) workloads launch hundreds to thousands of short-running GPU kernels per iteration. With GPU compute throughput growing rapidly, CPU-side launch latency of ker…
vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention
Ramya Prabhu, Ajay Nayak, Jayashree Mohan +2
PagedAttention is a popular approach for dynamic memory allocation in LLM serving systems. It enables on-demand allocation of GPU memory to mitigate KV cache fragmentation -- a phe…