high performance computing on graphics processing units: hgpu.org

Large Language Models (LLMs) have demonstrated strong capabilities in general-purpose code generation. However, generating the code which is deeply hardware-specific, architecture-aware, and performance-critical, especially for massively parallel GPUs, remains a complex challenge. In this work, we explore the use of LLMs for the automated generation and optimization of CUDA programs, with the goal of producing […]

The rapid growth of deep learning has driven exponential increases in model parameters and computational demands. NVIDIA GPUs and their CUDA-based software ecosystem provide robust support for parallel computing, significantly alleviating computational bottlenecks. Meanwhile, due to the cultivation of user programming habits and the high performance of GPUs, the CUDA ecosystem has established a dominant […]

GPUs are uniquely suited to accelerate (SQL) analytics workloads thanks to their massive compute parallelism and High Bandwidth Memory (HBM) — when datasets fit in the GPU HBM, performance is unparalleled. Unfortunately, GPU HBMs remain typically small when compared with lower-bandwidth CPU main memory. Besides brute-force scaling across many GPUs, current solutions to accelerate queries […]

This master thesis presents an implementation for enabling profile-guided optimizations (PGO) for mobile phone GPUs. PGO is an optimization technique that uses runtime profiling data, like block frequency and function call frequency, to guide compiler optimizations. The implementation is done by adapting the existing PGO infrastructure in LLVM to serve the architectural differences between CPUs […]

Machine learning potentials (MLPs) have advanced rapidly and show great promise to transform molecular dynamics (MD) simulations. However, most existing software tools are tied to specific MLP architectures, lack integration with standard MD packages, or are not parallelizable across GPUs. To address these challenges, we present chemtrain-deploy, a framework that enables model-agnostic deployment of MLPs […]

In this thesis, we address the challenge of performance portability in heterogeneous computing environments. Performance portability refers to the ability of an application to maintain high performance on multiple platforms without requiring extensive manual tuning for each system. Traditional containers fall short in this regard as they prioritize portability at the expense of architecture-specific optimizations. […]

As multicore vector processors improve in computational and memory performance, running SIMT (Single Instruction Multiple Threads) programs on CPUs has become increasingly appealing, potentially eliminating the need for dedicated GPU hardware. SYCL is a royalty-free cross-platform C++ programming model for heterogeneous computing that implements the SIMT model and provides a path to run GPU programs […]

Owing to the huge success of generative artificial intelligence (AI), large language models (LLMs) have emerged as a core subclass, underpinning applications such as question answering, text generation, and code completion. While fine-tuning these models on domain-specific data can yield significant performance gains, it also poses daunting computational challenges, especially for researchers and small organizations […]

Performing binary search on a sorted dense array is a widely used baseline when benchmarking sophisticated index structures: It is simple, fast to build, and indexes the dataset with minimal memory footprint. However, the popular opinion is that it cannot compete with sophisticated indexes in terms of lookup performance, and hence, should not actually be […]

GPU computing is embracing weak memory concurrency for performance improvement. However, compared to CPUs, modern GPUs provide more fine-grained concurrency features such as scopes, have additional properties like divergence, and thereby follow different weak memory consistency models. These features and properties make concurrent programming on GPUs more complex and error-prone. To this end, we present […]

Batched kernels with memory allocations is a common pattern in HPC, appearing in multi-dimensional FFTs, neural networks processing, or split computation of numerical operators. Its efficient support is especially complex on GPU where memory per work-item is limited and dynamic memory allocations are challenging. This study investigates whether the native abstractions of SYCL can support […]

This work examines latency, throughput, and other metrics when performing inference on confidential GPUs. We explore different traffic patterns and scheduling strategies using a single Virtual Machine with one NVIDIA H100 GPU, to perform relaxed batch inferences on multiple Large Language Models (LLMs), operating under the constraint of swapping models in and out of memory, […]