Related papers: Efficient Resource Sharing Through GPU Virtualizat…
The rise of LLMs has driven demand for private serverless deployments, characterized by moderate-sized models and infrequent requests. While existing serverless solutions follow exclusive GPU allocation, we take a step back to explore…
Over the last couple of years it has been realized that the vast computational power of graphics processing units (GPUs) could be harvested for purposes other than the video game industry. This power, which at least nominally exceeds that…
Modern GPUs increasingly rely on specialized and asynchronous hardware units to deliver high performance. Yet these units are often underutilized because today's GPU software stacks still organize programming and execution around a…
Large-scale parallel numerical simulations are essential for a wide range of engineering problems that involve complex, coupled physical processes interacting across a broad range of spatial and temporal scales. The data structures involved…
Efficient utilization of computing resources in a Kubernetes cluster is often constrained by the uneven distribution of pods with similar usage patterns. This paper presents a novel scheduling strategy designed to optimize the…
Fine-grained workload and resource balancing is the key to high performance for regular and irregular computations on the GPUs. In this dissertation, we conduct an extensive survey of existing load-balancing techniques to build an…
Traditionally, HPC workloads have been deployed in bare-metal clusters; but the advances in virtualization have led the pathway for these workloads to be deployed in virtualized clusters. However, HPC cluster administrators/providers still…
To support growing massive parallelism, functional components and also the capabilities of current processors are changing and continue to do so. Todays computers are built upon multiple processing cores and run applications consisting of a…
Diffusion models have achieved remarkable progress in high-fidelity image, video, and audio generation, yet inference remains computationally expensive. Nevertheless, current diffusion acceleration methods based on distributed parallelism…
Singular Value Decomposition (SVD) is a fundamental matrix factorization technique in linear algebra, widely applied in numerous matrix-related problems. However, traditional SVD approaches are hindered by slow panel factorization and…
Grid technologies aim at enabling a coordinated resource-sharing and problem-solving capabilities over local and wide area networks and span locations, organizations, machine architectures and software boundaries. The heterogeneity of…
Process mapping asks to assign vertices of a task graph to processing elements of a supercomputer such that the computational workload is balanced while the communication cost is minimized. Motivated by the recent success of GPU-based graph…
The future of computation is the Graphical Processing Unit, i.e. the GPU. The promise that the graphics cards have shown in the field of image processing and accelerated rendering of 3D scenes, and the computational capability that these…
GPUs have significantly accelerated first-order methods for large-scale optimization, especially in continuous optimization. However, this success has not transferred cleanly to problems with discrete variables, combinatorial structure, and…
With CPU scaling slowing down in today's data centers, more functionalities are being offloaded from the CPU to auxiliary devices. One such device is the SmartNIC, which is being increasingly adopted in data centers. In today's cloud…
In the fusion community, the use of high performance computing (HPC) has been mostly dominated by heavy-duty plasma simulations, such as those based on particle-in-cell and gyrokinetic codes. However, there has been a growing interest in…
This paper presents efforts to improve the hierarchical parallelism of a two scale simulation code. Two methods to improve the GPU parallel performance were developed and compared. The first used the NVIDIA Multi-Process Service and the…
Large-scale GPU traces play a critical role in identifying performance bottlenecks within heterogeneous High-Performance Computing (HPC) architectures. However, the sheer volume and complexity of a single trace of data make performance…
Matrix decompositions are ubiquitous in machine learning, including applications in dimensionality reduction, data compression and deep learning algorithms. Typical solutions for matrix decompositions have polynomial complexity which…
This paper provided empirical knowledge of the user experience for using collaborative visualization in a distributed asymmetrical setting through controlled user studies. With the ability to access various computing devices, such as…