English
Related papers

Related papers: Comparative evaluation of bandwidth-bound applicat…

200 papers

We investigate the performance characteristics of a numerically enhanced scalar product (dot) kernel loop that uses the Kahan algorithm to compensate for numerical errors, and describe efficient SIMD-vectorized implementations on recent…

Performance · Computer Science 2018-07-09 Johannes Hofmann , Dietmar Fey , Michael Riedmann , Jan Eitzinger , Georg Hager , Gerhard Wellein

The performance of discrete general purpose graphics processing units (GPGPUs) has been improving at a rapid pace. The PCIe interconnect that controls the communication of data between the system host memory and the GPU has not improved as…

Computational Physics · Physics 2019-05-15 Connor Kenyon , Glenn Volkema , Gaurav Khanna

Nowadays, latency-critical, high-performance applications are parallelized even on power-constrained client systems to improve performance. However, an important scenario of fine-grained tasking on simultaneous multithreading CPU cores in…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-10-03 Denis Los , Igor Petushkov

The balance metric is a simple approach to estimate the performance of bandwidth-limited loop kernels. However, applying the method to in-cache situations and modern multi-core architectures yields unsatisfactory results. This paper…

Performance · Computer Science 2009-10-27 Jan Treibig , Georg Hager , Gerhard Wellein

The AIPC concept is gaining popularity, and more and more hybrid CPUs will be running AI models on client devices. However, the current AI inference framework overlooks the imbalanced hardware capability of hybrid CPUs, leading to low…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-12-02 Luo Yu , Liu Yucheng , Shen Haihao

Exascale computing will get mankind closer to solving important social, scientific and engineering problems. Due to high prototyping costs, High Performance Computing (HPC) system architects make use of simulation models for design space…

Performance · Computer Science 2018-03-28 Alexandra Ferreron , Radhika Jagtap , Sascha Bischoff , Roxana Rusitoru

Specialized accelerators dominate AI workloads, but CPUs remain critical for orchestrating these accelerators and running datacenter services. As a result, CPU performance increasingly shapes end-to-end system efficiency, making it…

Hardware Architecture · Computer Science 2026-05-08 Ruihao Li , Andrew Jacob , Neeraja J. Yadwadkar , Lizy K. John

In this work, we first characterize the hybrid execution patterns of GCNs on Intel Xeon CPU. Guided by the characterization, we design a GCN accelerator, HyGCN, using a hybrid architecture to efficiently perform GCNs. Specifically, first,…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-01-09 Mingyu Yan , Lei Deng , Xing Hu , Ling Liang , Yujing Feng , Xiaochun Ye , Zhimin Zhang , Dongrui Fan , Yuan Xie

Here, we test the performance and scalability of fully-asynchronous, best-effort communication on existing, commercially-available HPC hardware. A first set of experiments tested whether best-effort communication strategies can benefit…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-10-10 Matthew Andres Moreno , Charles Ofria

Modern HPC systems are increasingly relying on greater core counts and wider vector registers. Thus, applications need to be adapted to fully utilize these hardware capabilities. One class of applications that can benefit from this increase…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-06-16 James Vance , Zhen-Hao Xu , Nikita Tretyakov , Torsten Stuehn , Markus Rampp , Sebastian Eibl , Christoph Junghans , André Brinkmann

Manycores are consolidating in HPC community as a way of improving performance while keeping power efficiency. Knights Landing is the recently released second generation of Intel Xeon Phi architecture. While optimizing applications on CPUs,…

Distributed, Parallel, and Cluster Computing · Computer Science 2018-11-06 Enzo Rucci , Armando De Giusti , Marcelo Naiouf

End-to-end (E2E) artificial intelligence (AI) pipelines are composed of several stages including data preprocessing, data ingestion, defining and training the model, hyperparameter optimization, deployment, inference, postprocessing,…

With multi-core processors a ubiquitous building block of modern supercomputers, it is now past time to enable applications to embrace these developments in processor design. To achieve exascale performance, applications will need ways of…

Distributed, Parallel, and Cluster Computing · Computer Science 2012-08-13 Michele Weiland , Lawrence Mitchell , Gerard Gorman , Stephan Kramer , Mark Parsons , James Southern

Fully provisioned Message Passing Interface (MPI) parallelism achieves near-optimal wall-clock time for Computational Fluid Dynamics (CFD) solvers. This work addresses a complementary question for shared, cloud-managed clusters: can…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-03-25 Tianfang Xie

The emergence of new, off-path smart network cards (SmartNICs), known generally as Data Processing Units (DPU), has opened a wide range of research opportunities. Of particular interest is the use of these and related devices in tandem with…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-04 Nathan Tibbetts , Sifat Ibtisum , Satish Puri

With the growing number of data-intensive workloads, GPU, which is the state-of-the-art single-instruction-multiple-thread (SIMT) processor, is hindered by the memory bandwidth wall. To alleviate this bottleneck, previously proposed…

Hardware Architecture · Computer Science 2021-03-12 Xinfeng Xie , Peng Gu , Yufei Ding , Dimin Niu , Hongzhong Zheng , Yuan Xie

The server central processing unit (CPU) market continues to exhibit robust demand due to the rising global need for computing power. Against this backdrop, CPU benchmark performance prediction is crucial for architecture designers. It…

Performance · Computer Science 2024-10-29 Xiaoman Liu

Performance modeling of parallel applications on multicore computers remains a challenge in computational co-design due to the complex design of multicore processors including private and shared memory hierarchies. We present a Scalable…

The runtime of a Lattice QCD simulation is dominated by a small kernel, which calculates the product of a vector by a sparse matrix known as the "Dslash" operator. Therefore, this kernel is frequently optimized for various HPC…

Distributed, Parallel, and Cluster Computing · Computer Science 2014-09-05 O. Kaczmarek , C. Schmidt , P. Steinbrecher , Swagato Mukherjee , M. Wagner

B-spline based orbital representations are widely used in Quantum Monte Carlo (QMC) simulations of solids, historically taking as much as 50% of the total run time. Random accesses to a large four-dimensional array make it challenging to…

Distributed, Parallel, and Cluster Computing · Computer Science 2017-08-10 Amrita Mathuriya , Ye Luo , Anouar Benali , Luke Shulenburger , Jeongnim Kim