This article explains how to design and implement a Kubernetes extension that allows multiple AI workloads to share a single Nvidia GPU by defining new extended resources, using a scheduler extender and a device plugin, and provides deployment steps, demos, and open‑source references.
This article provides a step‑by‑step guide on installing nvidia‑docker2, configuring the NVIDIA container runtime, setting up the GPU device plugin on a Kubernetes cluster, and deploying a GPU‑enabled pod to demonstrate proper GPU scheduling and usage.
The article reviews the two main development routes for artificial intelligence chips—traditional architectures that enhance CPU, GPU, and FPGA performance and emerging neuromorphic designs like IBM TrueNorth—detailing their structures, advantages, limitations, and industry adoption.
This article details the practical steps to install nvidia-docker2, configure Docker’s runtime, enable GPU support via Kubernetes device plugins, and verify GPU scheduling on a large Kubernetes cluster, providing code snippets and best‑practice recommendations for production environments.
This tutorial walks through installing NVIDIA‑Docker 2.0, configuring Docker’s runtime, deploying the NVIDIA device plugin on a Kubernetes 1.9 cluster, and testing GPU‑enabled pods, highlighting the advantages over the legacy nvidia‑docker 1.0 approach.
This article describes the architecture, components, and deployment strategies of the 58 deep learning online prediction service, covering TensorFlow‑Serving, custom model serving, traffic forwarding, load balancing, GPU configuration, resource monitoring, and the supporting web management platform.
The article presents WHLSL, a new Web‑compatible high‑level shading language inspired by HLSL, detailing its safety‑first design, language requirements, syntax features, standard‑library concepts, compilation stages, and how it maps to Metal, Direct3D 12, and SPIR‑V for modern WebGPU applications.
The 52nd TOP500 list released in November 2018 shows US DOE machines dominating the top ten, highlights performance gains of Summit and Sierra, details the hardware of leading supercomputers worldwide, and analyzes country shares, manufacturers, accelerator usage, interconnect technologies, and energy‑efficiency rankings such as Green500 and HPCG.
This article explains what a GPU (Graphics Processing Unit) is, why it differs from a CPU, how it is used through graphics and compute APIs, and explores GPU virtualization techniques such as virtual GPUs, passthrough, and vGPU architectures.
The OpenStack Cyborg project provides a generic framework that lets cloud platforms discover, schedule, and control proprietary accelerators such as GPUs, FPGAs, ASICs and SoCs, solving resource waste in AI, NFV, edge and HPC workloads.
Deep neural networks, propelled by breakthroughs such as AlexNet and advances in GPU and TPU hardware, are rapidly moving from academic research into diverse applications—including earthquake prediction, medical imaging, and autonomous driving—driving massive industry investment, new semiconductor designs, and intense competition among tech giants and startups.
The article explains how heterogeneous parallel computing—leveraging CPUs, GPUs, FPGAs and other specialized units—addresses the performance limits of traditional serial programming by distributing tasks across diverse hardware, detailing its concepts, architectures, development models, and relevance to AI and cloud workloads.
This article explains the historical development of graphics processing units, their internal structure, rendering pipeline, and how GPUs shifted graphics workloads from CPUs to specialized parallel hardware, highlighting key concepts such as vertex shaders, pixel shaders, SIMD architectures, and performance growth.
This article explains how browsers can leverage GPU acceleration via WebGL and GLSL to perform general‑purpose calculations, compares CPU and GPU architectures, demonstrates step‑by‑step shader programming, data encoding into RGBA, and discusses precision limits and performance gains.
This article provides a comprehensive overview of high‑performance computing, covering its rapid market growth, definition, classification into high‑throughput and distributed computing, key hardware components such as CPUs, GPUs, memory types, networking technologies like InfiniBand, performance metrics, benchmarking tools, and parallel file systems.
The article explains how CPUs, despite being usable for early cryptocurrency mining, are inefficient due to limited parallelism and general‑purpose design, while GPUs and ASICs provide massive parallel integer processing that makes them far more effective for modern proof‑of‑work algorithms.
This guide walks you through installing Anaconda, creating a TensorFlow virtual environment on Windows, configuring CPU and GPU versions, and implementing both a basic softmax regression and a deep convolutional neural network for MNIST digit recognition, complete with code snippets, training tips, and visualization tools.
This beginner‑friendly TensorFlow tutorial by Chen Yidong walks readers through Windows environment setup, explains TensorFlow’s graph‑execution model, and demonstrates both softmax linear regression and a deep convolutional neural network for MNIST, while also covering utility scripts, TensorBoard visualization, and CPU/GPU or multi‑GPU deployment.
This guide answers common beginner questions about deep learning, covering the essential hardware (especially GPUs and why Nvidia dominates), recommended software libraries, the choice between dynamic and static computation graphs, production considerations, required coding background, and how small datasets can still yield powerful models.
The article explains why general‑purpose CPUs can no longer meet deep‑learning demands due to intrinsic scaling limits and memory‑bandwidth bottlenecks, and surveys how heterogeneous accelerators—GPUs, FPGAs, ASICs and emerging AI processors with high‑bandwidth memory—provide specialized, high‑parallelism, power‑efficient solutions for both cloud and edge workloads.
The article analyzes the limitations of general‑purpose CPUs for deep‑learning workloads, explains how semiconductor scaling and memory‑bandwidth constraints drive the shift toward specialized heterogeneous processors such as GPUs, FPGAs, and ASICs, and discusses the design trade‑offs of embedded versus cloud AI accelerators.
This article provides a comprehensive, hands‑on tutorial for setting up a CentOS 6.8 server with NVIDIA GPUs, installing Java 8, GCC 4.9.4, Python 2.7, NVIDIA drivers, CUDA 8.0, cuDNN 6.0, Bazel, and finally building and verifying TensorFlow 1.3.0 from source, while documenting common pitfalls and fixes.
This article explores WebGL rendering fundamentals, compares vertex and fragment shader workloads, and presents practical optimization techniques such as moving calculations to shaders, reducing uniform updates, and minimizing program switches to achieve smoother frame rates.
This article compares NVIDIA's Tesla and GeForce GPU families, detailing their target markets, design differences, core architectures, double‑precision performance, ECC support, memory bandwidth, interface options, software and OS compatibility, power efficiency, and management features to help readers choose the right GPU for HPC or gaming workloads.
This guide explains how to identify performance bottlenecks in WebGL applications by examining CPU and GPU workloads, using tools such as stats.js, Chrome DevTools, Web Tracing Framework, and offers concrete optimization strategies for shaders, draw calls, state changes, and resource management.
This article explains the fundamentals of web animation performance, examines a complex canvas‑based weather animation case, introduces tools like Chrome's FPS meter and stats.js, and details browser rendering pipelines, CPU/GPU roles, and practical optimization techniques for smoother front‑end experiences.
This article reviews how GPU memory bandwidth, architecture, and price affect deep‑learning workloads, compares popular NVIDIA models and alternatives, discusses multi‑GPU scaling, and provides practical guidance for selecting the most suitable GPU for various budgets and use cases.
This article explains how GPU acceleration can keep web pages within the 16 ms frame budget, details the browser rendering pipeline, shows how to use Chrome DevTools to inspect composited layers, and outlines practical steps and pitfalls for enabling GPU‑accelerated animations.
Hardware acceleration in Android shifts intensive floating‑point UI work from the CPU to the GPU by building DisplayLists on the CPU and rasterizing them on the GPU, allowing parallel processing, selective redraw of unchanged elements, and significantly higher frame rates for animations and complex graphics.
This article provides a detailed, step‑by‑step guide for iOS developers on measuring and optimizing app performance across CPU usage, GPU‑related stutter, network conditions, and energy consumption using built‑in Instruments, Activity Monitor, Network Link Conditioner, and Charles.
The article announces the open‑source release of kaldi‑ctc, a GPU‑accelerated CTC‑based end‑to‑end speech recognition toolkit built on Kaldi, warp‑ctc and cuDNN, highlighting its 5‑6× training speedup, real‑time decoding factor of 0.02, and performance comparisons on the LibriSpeech corpus.
This article explains how to quantify Android app frame rates, understand FPS concepts, calculate frame timing, and apply three practical methods—including GPU rendering profiling, adb gfxinfo, and SurfaceFlinger latency commands—to monitor and improve smoothness across multiple devices without rooting or API restrictions.
This article provides a light‑hearted yet thorough overview of deep learning, covering neural network fundamentals, layer construction, back‑propagation, ResNet shortcuts, encoder‑decoder structures, PU‑learning for unlabeled data, GPU acceleration, and practical advice on data size, frameworks, and deployment in financial scenarios.
This article provides a humorous yet informative, non‑mathematical guide to deep learning, covering neural network basics, layer addition, training methods, back‑propagation, unsupervised pre‑training, regularization, ResNet shortcuts, GPU computation, framework choices, and practical advice for applying deep learning to industrial data.
This article describes how Yahoo integrated deep learning into its massive Hadoop ecosystem by adding GPU nodes, using YARN and Spark to run Caffe at scale, and presents performance results on AlexNet and GoogLeNet alongside open‑source contributions.
Looking ahead, programmers may see GPUs supplant CPUs, databases handle complex analytics, JavaScript dominate browsers and servers, Android power every device, the Internet of Things expand, open‑source ecosystems seek sustainable funding, CMS tools become richer, plugin‑centric development replace monolithic coding, command‑line interfaces endure, education strives for true simplification, outsourcing battles automation, and managers risk growing technically illiterate.
