Related papers: jaxsgp4: GPU-accelerated mega-constellation propag…
The emergence of Large Language Models (LLMs) has necessitated the adoption of distributed training techniques, involving the deployment of thousands of GPUs to train a single model. Unfortunately, the efficiency of large-scale distributed…
Sparse data structures are commonly used in neural networks to reduce the memory footprint. These data structures are compact but cause irregularities such as random memory accesses, which prevent efficient use of the memory hierarchy. GPUs…
Sparse linear algebra kernels play a critical role in numerous applications, covering from exascale scientific simulation to large-scale data analytics. Offloading linear algebra kernels on one GPU will no longer be viable in these…
Strong gravitational lensing is a powerful probe of cosmology and the dark matter distribution. Efficient lensing software is already a necessity to fully use its potential and the performance demands will only increase with the upcoming…
The prevailing assumption holds that on-orbit servicing (OOS) of mega-constellations is infeasible due to prohibitive fuel consumption incurred by multiple rendezvous maneuvers across vast and dispersed satellite populations. To address…
Incoherent dedispersion is a computationally intensive problem that appears frequently in pulsar and transient astronomy. For current and future transient pipelines, dedispersion can dominate the total execution time, meaning its…
Exploiting sparsity underlying neural networks has become one of the most potential methodologies to reduce the memory footprint, I/O cost, and computation workloads during inference. And the degree of sparsity one can exploit has become…
We introduce a parallel GPU implementation of the Simple Linear Iterative Clustering (SLIC) superpixel segmentation. Using a single graphic card, our implementation achieves speedups of up to $83\times$ from the standard sequential…
The JPEG compression format has been the standard for lossy image compression for over multiple decades, offering high compression rates at minor perceptual loss in image quality. For GPU-accelerated computer vision and deep learning tasks,…
We describe source code level parallelization for the {\tt kira} direct gravitational $N$-body integrator, the workhorse of the {\tt starlab} production environment for simulating dense stellar systems. The parallelization strategy, called…
Real-time remote sensing applications like search and rescue missions, military target detection, environmental monitoring, hazard prevention and other time-critical applications require onboard real time processing capabilities or…
Upcoming large scale telescope projects such as the Square Kilometre Array (SKA) will see high data rates and large data volumes; requiring tools that can analyse telescope event data quickly and accurately. In modern radio telescopes,…
Recent advances to algorithms for training spiking neural networks (SNNs) often leverage their unique dynamics. While backpropagation through time (BPTT) with surrogate gradients dominate the field, a rich landscape of alternatives can…
Low-latency detections of gravitational waves (GWs) are crucial to enable prompt follow-up observations to astrophysical transients by conventional telescopes. We have developed a low-latency pipeline using a technique called Summed…
Large-scale diffusion models such as FLUX (12B parameters) and Stable Diffusion 3 (8B parameters) require multi-GPU parallelism for efficient inference. Unified Sequence Parallelism (USP), which combines Ulysses and Ring attention…
A major bottleneck in scenario-based Sample Average Approximation (SAA) for stochastic programming (SP) is the cost of solving an exact second-stage problem for every scenario, especially when each scenario contains an NP-hard combinatorial…
The rapid development in computing technology has paved the way for directive-based programming models towards a principal role in maintaining software portability of performance-critical applications. Efforts on such models involve a least…
Gaussian processes (GPs) have become a common tool in astronomy for analysing time series data, particularly in exoplanet science and stellar astrophysics. However, choosing the appropriate covariance structure for a GP model remains a…
In this paper we describe and demonstrate a C++ code written to determine the trajectory of particles traversing oriented single crystals and a CUDA code written to evaluate the radiation spectra from charged particles with arbitrary…
In computational science and data analytics, many workloads involve irregular and sparse computations that are inherently difficult to optimize for modern hardware. A key kernel is Sparse General Matrix-Matrix Multiplication (SpGEMM), which…