English
Related papers

Related papers: Benchmarking and Dissecting the Nvidia Hopper GPU …

200 papers

NVIDIA has been making steady progress in increasing the compute performance of its GPUs, resulting in order of magnitude compute throughput improvements over the years. With several models of GPUs coexisting in many deployments, the…

Distributed, Parallel, and Cluster Computing · Computer Science 2022-07-12 Igor Sfiligoi , David Schultz , Frank Würthwein , Benedikt Riedel , Dmitry Y. Mishin

Many architects believe that major improvements in cost-energy-performance must now come from domain-specific hardware. This paper evaluates a custom ASIC---called a Tensor Processing Unit (TPU)---deployed in datacenters since 2015 that…

Modern graphics computing units (GPUs) are designed and optimized to perform highly parallel numerical calculations. This parallelism has enabled (and promises) significant advantages, both in terms of energy performance and calculation. In…

Hardware Architecture · Computer Science 2021-10-26 Quentin Gallouédec

Fast training of large machine learning models requires distributed training on AI clusters consisting of thousands of GPUs. The efficiency of distributed training crucially depends on the efficiency of the network interconnecting GPUs in…

Networking and Internet Architecture · Computer Science 2025-06-11 Erfan Nosrati , Majid Ghaderi

In this paper, we present a novel technique to search for hardware architectures of accelerators optimized for end-to-end training of deep neural networks (DNNs). Our approach addresses both single-device and distributed pipeline and tensor…

Hardware Architecture · Computer Science 2024-04-24 Muhammad Adnan , Amar Phanishayee , Janardhan Kulkarni , Prashant J. Nair , Divya Mahajan

Fully Homomorphic Encryption (FHE) enables computation directly on encrypted data but incurs massive computational and memory overheads, often exceeding plaintext execution by several orders of magnitude. While custom ASIC accelerators can…

Massively multicore processors, such as Graphics Processing Units (GPUs), provide, at a comparable price, a one order of magnitude higher peak performance than traditional CPUs. This drop in the cost of computation, as any…

Distributed, Parallel, and Cluster Computing · Computer Science 2016-11-18 Samer Al-Kiswany , Abdullah Gharaibeh , Matei Ripeanu

This paper explores the impact of simulator accuracy on architecture design decisions in the general-purpose graphics processing unit (GPGPU) space. We perform a detailed, quantitative analysis of the most popular publicly available GPU…

Hardware Architecture · Computer Science 2020-06-04 Mahmoud Khairy , Jain Akshay , Tor Aamodt , Timothy G. Rogers

Deep learning kernels exhibit predictable memory accesses and compute patterns, making GPUs' parallel architecture well-suited for their execution. Software and runtime systems for GPUs are optimized to better utilize the stream…

Machine Learning · Computer Science 2024-12-13 Seonho Lee , Amar Phanishayee , Divya Mahajan

The simulation of the two-dimensional Ising model is used as a benchmark to show the computational capabilities of Graphic Processing Units (GPUs). The rich programming environment now available on GPUs and flexible hardware capabilities…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-08-26 Joshua Romero , Mauro Bisson , Massimiliano Fatica , Massimo Bernaschi

Accurate hardware performance models are critical to efficient code generation. They can be used by compilers to make heuristic decisions, by superoptimizers as a minimization objective, or by autotuners to find an optimal configuration for…

This document presents a vision for a novel AI infrastructure design that has been initially validated through inference simulations on state-of-the-art large language models. Advancements in deep learning and specialized hardware have…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-17 Jiamin Li , Lei Qu , Tao Zhang , Grigory Chirkov , Shuotao Xu , Peng Cheng , Lidong Zhou

The rapid scaling of large language models (LLMs) has unveiled critical limitations in current hardware architectures, including constraints in memory capacity, computational efficiency, and interconnection bandwidth. DeepSeek-V3, trained…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-24 Chenggang Zhao , Chengqi Deng , Chong Ruan , Damai Dai , Huazuo Gao , Jiashi Li , Liyue Zhang , Panpan Huang , Shangyan Zhou , Shirong Ma , Wenfeng Liang , Ying He , Yuqing Wang , Yuxuan Liu , Y. X. Wei

This work examines the performance of leading-edge systems designed for machine learning computing, including the NVIDIA DGX-2, Amazon Web Services (AWS) P3, IBM Power System Accelerated Compute Server AC922, and a consumer-grade Exxact…

Performance · Computer Science 2019-10-03 Yihui Ren , Shinjae Yoo , Adolfy Hoisie

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…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-03-16 Petr Samoldekin , Christian Schulz , Henning Woydt

Learning an algorithm from examples is a fundamental problem that has been widely studied. Recently it has been addressed using neural networks, in particular by Neural Turing Machines (NTMs). These are fully differentiable computers that…

Machine Learning · Computer Science 2016-03-16 Łukasz Kaiser , Ilya Sutskever

In recent times, the trend in very large scale integration (VLSI) industry is multi-dimensional, for example, reduction of energy consumption, occupancy of less space, precise result, less power dissipation, faster response. To meet these…

Machine Learning · Computer Science 2021-07-02 Gaurab Bhattacharya

As AI chips incorporate numerous parallelized cores to scale deep learning (DL) computing, inter-core communication is enabled recently by employing high-bandwidth and low-latency interconnect links on the chip (e.g., Graphcore IPU). It…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-09-25 Yiqi Liu , Yuqi Xue , Yu Cheng , Lingxiao Ma , Ziming Miao , Jilong Xue , Jian Huang

Collocating deep learning training tasks improves GPU utilization but risks resource contention, severe slowdowns, and out-of-memory (OOM) failures. Accurate memory estimation is essential for robust collocation, and GPU utilization…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-29 Ehsan Yousefzadeh-Asl-Miandoab , Reza Karimzadeh , Danyal Yorulmaz , Bulat Ibragimov , Pınar Tözün

GPU architectures have become popular for executing general-purpose programs. Their many-core architecture supports a large number of threads that run concurrently to hide the latency among dependent instructions. In modern GPU…

Hardware Architecture · Computer Science 2024-01-19 Rodrigo Huerta , Mojtaba Abaie Shoushtary , Antonio González