T4는 세계를 선도하는 엔터프라이즈 서버에 GPU 가속화를 제공합니다.
NVIDIA T4 엔터프라이즈 GPU와 CUDA-X 가속화 라이브러리는 오늘날의 최신 데이터센터에 맞게 설계된 메인스트림 서버를 가동시킵니다. 이들 서버는 NGC-Ready 검증을 통해 가속화 워크로드의 전체 범위에서 뛰어난 성능을 보여줍니다. NVIDIA NGC 지원 서비스 를 통한 지원 옵션으로 안심할 수 있습니다. NVIDIA T4를 통해 IT 팀은 세계 최고의 서버를 갖춘 표준 GPU 가속화 데이터센터 인프라를 구축할 수 있습니다.
최대 33% 개선된 가상 데스크톱 인프라(VDI)* 성능
최대 35배 빠른 머신 러닝 교육
최대 10배 빠른 딥 러닝 교육
최대 40배 빠른 딥 러닝 추론 인사이트 확보
구성 정보
딥 러닝은 근래에 AI 시장에 도입된 반면, 머신 러닝은 수십 년 전부터 기업들이 데이터에서 인사이트를 얻는 데 사용해 왔습니다. NVIDIA T4 데이터센터 GPU는 GPU 가속화 데이터 준비와 머신 러닝을 위한 일련의 오픈 소스 라이브러리인 CUDA-X AI 소프트웨어 가속화 라이브러리 RAPIDS를 사용해 이러한 머신 러닝 기법을 가속할 수 있습니다. T4 GPU는 Python 등의 익숙한 개발 도구 사용 시 XGBoost, PCA, K-평균 k-NN, DBScan, tSVD 등의 알고리즘을 포함한 CPU 전용 서버에 비해 머신 러닝을 최대 35배 더 가속할 수 있습니다.
NVIDIA T4는 AI 워크로드를 위한 다중 정밀도 컴퓨팅을 갖춘 획기적인 NVIDIA Turing™ Tensor 코어 기술을 기반으로 합니다. FP32, FP16, INT8에서 INT4까지 획기적인 성능을 보여주는 T4는 CPU보다 최대 40배 더 높은 추론 성능을 발휘합니다. 트레이닝 시에는 2개의 T4 GPU가 있는 단일 서버가 9개의 듀얼 소켓 CPU 전용 서버를 대체합니다. 개발자들은 NVIDIA CUDA-X AI 소프트웨어 라이브러리와 모든 AI 프레임워크와의 통합을 통해 직접 Turing Tensor 코어의 힘을 일깨울 수 있습니다. CUDA 위에 구축된 NVIDIA의 병렬 프로그래밍 모델 CUDA-X 라이브러리는 인공 지능, 자동화 기계, 고성능 컴퓨팅, 그래픽의 특정 컴퓨팅 요건에 맞춰 최적화합니다.
NVIDIA T4를 탑재한 주류 서버는 가상 GPU 소프트웨어를 위해 당사 파트너가 인증하며 사용자가 장소에 관계없이 모든 장치에서 고품질 가상 데스크톱을 경험할 수 있게 합니다. 지식 근로자 및 모바일 전문가부터 디자이너 및 엔지니어에 이르기까지, T4를 NVIDIA 가상 GPU 소프트웨어(NVIDIA GRID® 가상 PC(GRID vPC)와 NVIDIA Quadro® 가상 데이터센터 워크스테이션(Quadro vDWS)과 결합하면 CPU 전용 VDI 환경에 비해 최대 33% 더 개선된 성능을 제공하며 VDI 경험을 가속할 수 있습니다.
세계적 수준의 성능을 발휘하도록 소프트웨어를 최적화하려면 운영 체제와 드라이버부터 라이브러리, 프레임워크, 애플리케이션까지 여러 가지 다양한 소프트웨어 계층을 깊이 이해해야 합니다. NGC 컨테이너는 사전 구축되며 GPU 컴퓨팅을 위해 고도로 최적화됩니다. 이들 컨테이너는 GPU 가속화 딥 러닝 프레임워크, RAPIDS 소프트웨어, NVIDIA TensorRT™, 그리고 600개 이상의 고성능 컴퓨팅 애플리케이션을 위한 NVIDIA의 전례 없는 전문 지식과 에코시스템 지원을 제공합니다. NVIDIA의 NGC-Ready 프로그램은 모든 서버 제작자가 자신의 서버에서 NGC 컨테이너를 검증하도록 지원합니다. NGC-Ready 서버는 데이터센터 관리자가 광범위한 가속화 애플리케이션에서 뛰어난 성능을 발휘하도록 신속하게 설정할 수 있다는 것을 알고 GPU 탑재 서버를 배포할 수 있도록 자신감을 심어줍니다.
VDI: GRID vPC tested on a server with 2x Intel Xeon Gold 6148 (20c, 2.4 GHz), GRID vPC with T4-1B (64 VMs), VMware ESXi 6.7, NVIDIA vGPU Software (410.91/412.16), Windows 10 (1803), 2 vCPUs, 4GB RAM, Resolution 1920x1080 resolution, Single Monitor, VMware Horizon 7.6 User experience was measured using an NVIDIA internal benchmarking tool which measured remoted frames running office productivity applications such as Microsoft PowerPoint, Word, Excel, Chrome, PDF viewing and video playback.
Machine Learning: CPU nodes (61 GB of memory, 8 vCPUs, 64-bit platform), Apache Spark. 200 GB CSV dataset; Data preparation includes joins, variable transformations. GPU Server Config: Dual-Socket Xeon E5-2698 v4@3.6GHz, 20 T4 GPUs on 5 nodes, each with 4 T4 GPUs. All run on InfiniBand network, CPU data for XGBoost and Data Conversion steps are estimated based on measured data for 20 CPU nodes, and reducing execution time by 60% to normalize for training on smaller data set on T4.
Deep Learning Training and Inference: GPU: Dual-Socket Xeon E5-2698 v4@3.6GHz. GPU Servers: 2xT4s for Training, 1xT4 for Inference, NGC 18.11-py3 Container with CUDA 10.0.130; NCCL 2.3.7, cuDNN 7.4.1.5; cuBLAS 10.0.130 | NVIDIA Driver: 384.145.
Cisco enables organizations and partners to embrace the latest innovations that drive digital transformation. We help organizations grow their businesses and respond to changing technology, customer, and market trends. For more than 30 years, Cisco has helped our customers build networks, automate, orchestrate, integrate, and digitize information technology (IT)-based products and services. Today, Cisco is helping lead the way by transforming businesses, governments, and cities worldwide with differentiated innovations like cloud computing, analytics, machine learning, security, collaboration, and artificial intelligence.
Learn More >
Dell empowers countries, communities, customers and people everywhere to use technology to realize their dreams. Customers trust us to deliver technology solutions that help them do and achieve more, whether they're at home, work, school or anywhere in their world.
Regardless of your HPC, AI, or VDI requirements, HPE has the best in class people, technology and partners to ensure you’re ready for today and tomorrows’ projects and challenges. We continuously innovate and deliver differentiated production-level solutions. Get faster insights with accelerated IT and solutions, using leading edge technologies and services from HPE.
Fujitsu is the leading Japanese information and communication technology (ICT) company, offering a full range of technology products, solutions, and services. Approximately 140,000 Fujitsu people support customers in more than 100 countries. We use our experience and the power of ICT to shape the future of society with our customers. Fujitsu Limited (TSE: 6702) reported consolidated revenues of 4.1 trillion yen (US $39 billion) for the fiscal year ended March 31, 2018.
Inspur is a leading intelligent computing solutions provider, ranked among the world’s top three server vendors. Inspur’s cutting-edge hardware products and designs are widely delivered and deployed in major data centers around the globe, serving important technology arenas like open computing, cloud, AI and deep learning. Inspur provides customers with customized IT infrastructures and AI solutions that are Tier 1 in quality and energy efficiency. Inspur’s products are optimized for applications and workloads built for data center environments.
At Lenovo, our goal is to be the most trusted data center partner – empowering customers’ Intelligent Transformation and solving humanity’s greatest challenges.
Sugon is a leading enterprise in the field of HPC, servers, storage, cloud and big data in China. Focus on HPC since 1993 and crown China HPC Top100 9 times in 10 years. sugon has mastered a series of core technologies of HPC, cooling Technology, Network Technology, Artificial Intelligence Technology and so on. Since 2006, strategic partner to NVIDIA. Having full spectrum GPU workstations and servers, inference server, training server, AI/RTX workstation.
Supermicro (SMCI), a leading innovator in high-performance, high-efficiency server technology, is a premier provider of advanced server Building Block Solutions® for Data Center, Cloud Computing, Enterprise IT, Hadoop/Big Data, AI/Deep Learning, HPC and Embedded Systems worldwide. Supermicro is committed to protecting the environment through its “We Keep IT Green®” initiative and provides customers with the most energy-efficient, environmentally-friendly solutions available on the market.
ASRock Rack Inc., established in 2013, specializes in the field of cloud computing server hardware. While inheriting ASRock’s design concepts, “Creativity, Consideration, Cost-effectiveness,” the company is dedicated to bringing the server industry out of the box thinking with the passion to innovate. Leveraged by ASRock’s growing momentum and distribution channels, this young and vibrant company targets the booming market of cloud computing and commits to serving the market with user friendly and eco-friendly "do-it-yourself" server hardware, featuring flexible and reliable products.
GIGABYTE has released and continues to develop a range of high performance computing systems. From strict component selection to a focus on design that meets market requirements, GIGABYTE demands and achieves the high product quality standards. GIGABYTE HPC systems are not only renowned for quality and reliability, but also the flexibility offered that allows scientists to meet simulation and analytics challenges. As a result of close cooperation with NVIDIA, GIGABYTE offers a comprehensive portfolio of systems compatible with NVIDIA Tesla and GRID accelerators.
Quanta Cloud Technology (QCT) is a global datacenter solution provider that combines the efficiency of hyperscale hardware with infrastructure software from a diversity of industry leaders to solve next-generation datacenter design and operation challenges. QCT serves cloud service providers, telecoms and enterprises running public, hybrid and private clouds. Product lines include hyper-converged and software-defined datacenter solutions as well as servers, storage, switches, integrated racks with a diverse ecosystem of hardware component and software partners. QCT designs, manufactures, integrates and services cutting edge offerings via its own global network. The parent of QCT is Quanta Computer, Inc., a Fortune Global 500 corporation.
TYAN is a leading provider of server motherboards and server platforms. TYAN designs, manufactures, and markets advanced x86 and x86-64 server/workstation board technology, platforms, and server solution products. Its products are sold to OEMs, VARs, system integrators and resellers worldwide for a wide range of applications. TYAN enables its customers to be technology leaders by providing scalable, highly integrated, and reliable products for a wide range of applications such as server appliances and solutions for HPC, hyper-scale/data center, server storage, and security appliance markets.
Wiwynn is an innovative cloud IT infrastructure provider of high quality computing and storage products, as well as rack solutions for leading data centers. Wiwynn aggressively invests in next generation technologies for workload optimization and best TCO (Total Cost of Ownership). As an OCP (Open Compute Project) solution provider and platinum member, Wiwynn actively participates in advanced computing and storage system designs while constantly implementing the benefits of OCP into traditional data centers.
ASUS is a Taiwan-based, multinational computer hardware and consumer electronics company that was established in 1989. ASUS became widely known in North America when it revolutionized the PC industry in 2007 with its Eee PC™. Today, the company is pioneering new mobile trends with the ASUS ZenFone™ series, and it is rapidly developing virtual and augmented reality products as well as IOT devices and robotics technologies.