Related papers: Sustaining Exascale Performance: Lessons from HPL …
Efficiently serving Large Language Models (LLMs) requires selecting an optimal parallel execution plan, balancing computation, memory, and communication overhead. However, determining the best strategy is challenging due to varying…
The unabated growth in AI workload demands is driving the need for concerted advances in compute, memory, and interconnect performance. As traditional semiconductor scaling slows, high-speed interconnects have emerged as the new scaling…
The rapid growth of Internet-of-things (IoT) and artificial intelligence applications have called forth a new computing paradigm--edge computing. In this paper, we study the suitability of deploying FPGAs for edge computing from the…
The high-performance computing (HPC) landscape is undergoing rapid transformation, with an increasing emphasis on energy-efficient and heterogeneous computing environments. This comprehensive study extends our previous research on SYCL's…
We carry out a comparative performance study of multi-core CPUs, GPUs and Intel Xeon Phi (Many Integrated Core - MIC) with a microscopy image analysis application. We experimentally evaluate the performance of computing devices on core…
Large-scale GPU clusters are widely-used to speed up both latency-critical (online) and best-effort (offline) deep learning (DL) workloads. However, most DL clusters either dedicate each GPU to one workload or share workloads in time,…
Designing efficient and scalable sparse linear algebra kernels on modern multi-GPU based HPC systems is a daunting task due to significant irregular memory references and workload imbalance across the GPUs. This is particularly the case for…
FPGA-based heterogeneous architectures provide programmers with the ability to customize their hardware accelerators for flexible acceleration of many workloads. Nonetheless, such advantages come at the cost of sacrificing programmability.…
Gaussian processes are widely used in machine learning domains but remain computationally demanding, limiting their efficient scalability across emerging hardware platforms. The GPRat library addresses these challenges using the HPX…
Hardware-based acceleration is an extensive attempt to facilitate many computationally-intensive mathematics operations. This paper proposes an FPGA-based architecture to accelerate the convolution operation - a complex and expensive…
Graphics Processing Units (GPUs) support dynamic voltage and frequency scaling (DVFS) in order to balance computational performance and energy consumption. However, there still lacks simple and accurate performance estimation of a given GPU…
Interconnection is crucial for computing systems. However, the current interconnection performance between processors and devices, such as memory devices and accelerators, significantly lags behind their computing performance, severely…
Many-core accelerators, as represented by the XeonPhi coprocessors and GPGPUs, allow software to exploit spatial and temporal sharing of computing resources to improve the overall system performance. To unlock this performance potential…
Benchmarking and comparing performance of a scientific simulation across hardware platforms is a complex task. When the simulation in question is constructed with an asynchronous, many-task (AMT) runtime offloading work to GPUs, the task…
The deployment of machine learning (ML) models on microcontrollers (MCUs) is constrained by strict energy, latency, and memory requirements, particularly in battery-operated and real-time edge devices. While software-level optimizations…
Large-scale deep learning benefits from an emerging class of AI accelerators. Some of these accelerators' designs are general enough for compute-intensive applications beyond AI and Cloud TPU is one such example. In this paper, we…
LLM training is scaled up to 10Ks of GPUs by a mix of data-(DP) and model-parallel (MP) execution. Critical to achieving efficiency is tensor-parallel (TP; a form of MP) execution within tightly-coupled subsets of GPUs, referred to as a…
Hybrid parallelism underpins large-scale LLM training across tens of thousands of GPUs. At such scale, hardware failures on individual devices lead to performance skew across devices, diminishing overall training efficiency. Existing…
As we approach the Exascale era, it is important to verify that the existing frameworks and tools will still work at that scale. Moreover, public Cloud computing has been emerging as a viable solution for both prototyping and urgent…
Tabular data remain the predominant format for real-world applications. Yet, developing effective neural models for tabular data remains challenging due to heterogeneous feature types and complex interactions occurring at multiple scales.…