SOLUCIONES DE DEEP LEARNING ON-PREMISES

Infunda su Negocio con Inteligencia Artificial Basada en el Data Center de Deep Learning

MAXIMIZAR EL RENDIMIENTO.
MINIMIZAR COSTOS.

Los data centers acelerados por GPU ofrecen un rendimiento innovador con menos servidores y menos potencia, lo que resulta en una visión más rápida con costos significativamente más bajos. Entrene los modelos de deep learning más complejos para resolver sus mayores desafíos con las soluciones de deep learning de NVIDIA con las GPU NVIDIA® para data centers. Ideal para la formación e inferencia a escala de producción, el rendimiento líder mundial de NVIDIA acelera los marcos de deep learning más populares y más de 550 aplicaciones de computación de alto rendimiento (HPC).

SOLUCIONES DE DEEP LEARNING ON-PREMISES

Entrenamiento

Entrenamiento

Para impulsar la IA, los data centers deben poder entrenar grandes cantidades de datos. Esto requiere infraestructura con alta potencia de cómputo, capacidad de almacenamiento generosa, eficiencia energética y escalabilidad. Descubra cómo NVIDIA está acelerando todas las cargas de trabajo de capacitación de deep learning..

Inferencia

Inferencia

La implementación de servicios de inteligencia artificial a la velocidad del rayo requiere un alto rendimiento y una baja latencia. Las GPU NVIDIA brindan aumentos dramáticos en eficiencia, rendimiento y escalabilidad al el data center, a un costo menor. Explore los beneficios de la plataforma de inferencia de deep learning NVIDIA.

IMPULSAR CARGAS DE TRABAJO IA CON GPU NVIDIA

Desde la formación hasta la inferencia, NVIDIA se está asociando con los OEM para crear sistemas certificados acelerados por GPU para cada implementación de IA.

Explore what’s available, and discover the right fit for your business.
Dell
HPE
Lenovo
NVIDIA

Vea la gama completa de soluciones disponibles de todos nuestros socios.

VER CATÁLOGO COMPLETO
Acceda al Software Optimizado para GPU Desde su Escritorio

Acceda al Software Optimizado para GPU Desde su Escritorio

Ahora puede descargar todo el software de deep learning que necesita de NVIDIA NGC de forma gratuita. NGC proporciona acceso simple a un catálogo completo de herramientas de software optimizadas para GPU para deep learning y computación de alto rendimiento (HPC). Aproveche al máximo las GPU NVIDIA en el escritorio, en el data center y en el cloud.

Explore NGC 

Explore los Recursos para el Data Centerspan Acelerado por GPU

Obtenga más recursos en soluciones de inteligencia artificial locales, directamente a su bandeja de entrada..

REGÍSTRATE

Cisco On-Premises AI Platforms

Cisco Unified Computing System (UCS™) solution for AI helps bridge the gap between Data Scientists and IT, avoids infrastructure silos through unified management, and with its ecosystem partners delivers a consistent experience for AI workloads running on-prem or in the cloud. Cisco UCS offers a full range of servers with flexible NVIDIA GPU configurations to address the full lifecycle of AI workloads at scale, from model development, to training and inferencing.

Cisco UCS C480ML M5

TRAINING AND MODEL DEVELOPMENT

UCS C480ML M5

  • 4U with 8 x V100 SXM2 NVLink, 3TB memory, 182TB storage
  • Ease of management with Cisco UCS Manager/Intersight
  • Cisco Validated with Flashstack AI, Flexpod AI CVDs
Learn More 
Cisco UCS C240 M5

ENTERPRISE AI DEPLOYMENT

UCS C240 M5

  • 2U with up to 6 x T4
  • Ease of management with Cisco UCS Manager/Intersight
  • AI, ML, DL Training, Inference, VDI, IVA, Video Transcode GPU Workloads
Learn More 

Dell EMC On-Premises AI Platforms

Dell Technologies is at the forefront of AI, providing the technology that makes tomorrow possible, today. Dell uniquely provides an extensive portfolio of technologies spanning from workstations, servers, networking, storage, software and consulting services to create the HPC and data analytics solutions that underpin successful AI, machine learning and deep learning solutions. Dell has also invested to create a portfolio of AI/ML Ready Solutions, simplifying the IT infrastructure to provide customers faster, deeper insights.

Dell EMC Poweredge C4140

TRAINING AND MODEL DEVELOPMENT

PowerEdge C4140

  • 1U with 4 x V100
  • 4-way Fully Connected NVLink GPU configuration
  • Enterprise Systems Manage with iDRAC and Open Manage
Learn More 
Dell EMC Poweredge R740

ENTERPRISE AI DEPLOYMENT

PowerEdge R740

  • 2U with up to 6 x T4
  • Enterprise Systems Manage with IDRAC and Open Manage
Learn More 

Hewlett Packard Enterprise On-Premises AI Platforms

From edge to hybrid cloud, HPE and NVIDIA have created a leading portfolio of optimized AI solutions that deliver deeper insights to solve the world’s greatest challenges. Our solutions are differentiated with proven AI expertise, the largest deep learning ecosystem, and AI software frameworks. Together, we equip industries and customers with the tools and resources they need to implement AI and deep learning, including learning opportunities through online and instructor-led workshops, reference architectures, and benchmark-winning GPU-accelerated applications that enhance time to value. Through joint development, testing, research, and validation, HPE and NVIDIA deliver the best solutions purpose- built for AI to the market today.

HPE Apollo 6500

TRAINING AND MODEL DEVELOPMENT

Apollo 6500

  • Variety of GPU support up to 8 V100 PCIe/SXM2, T4, RTX6000/RTX8000 GPUs
  • Enterprise Systems Management with iLO and management software suite
  • Multi-CPU:GPU topologies: 4:1 or 8:1
Learn More 
HPE Proliant DL380

ENTERPRISE AI DEPLOYMENT

Proliant DL380

  • Variety of GPU support in 2U chassis
  • Max. 3x V100 or RTX6000/RTX8000, 5x RTX4000, 7x T4 support
  • Enterprise Systems Management with iLO and management software suite
Learn More 

Lenovo On-Premises AI Platforms

Lenovo has developed an optimized AI training solution, combining Lenovo ThinkSystem SR670 with Lenovo Intelligent Computing Orchestration (LiCO) software to give clients confidence in deploying AI infrastructure without fear of stranding investment.  The Lenovo AI Starter kit solution enables businesses to quickly start running AI workloads in the data center with ease of use and productivity benefits for Data Scientists, maximum infrastructure utilization for IT, and easily allows for growth as the number of users and workloads scale.

ThinkSystem SR670

TRAINING AND MODEL DEVELOPMENT

ThinkSystem SR670

  • 2U High-Performance system
  • with modular I/O flexibility to
  • optimize accelerated workloads
  • Available GPU configurations
  • 4x NVIDIA V100 /V100S 32GB (AI training)
  • 8x NVIDIA T4 (AI Inferencing, Visualization)
  • 4 x RTX GPU’s
Learn More 
ThinkSytem SR650

ENTERPRISE AI DEPLOYMENT

ThinkSystem SR650

  • 2U with up to 5 x T4
  • 3 x V100 support
  • LiCO (Lenovo Intelligent Computing Orchestration)
Learn More 

NVIDIA On-Premises AI Platforms

Developed to meet the demands of AI and analytics, NVIDIA® DGX™ Systems are built on the revolutionary NVIDIA Volta™ GPU platform. Combined with innovative GPU-optimized software and simplified management tools, these fully-integrated solutions deliver groundbreaking performance and results. NVIDIA DGX Systems are designed to give data scientists the most powerful tools for AI exploration—from your desk to the data center to the cloud. Experiment faster, train larger models and arrive at insights—all starting on day one.

NVIDIA DGX-1

TRAINING, INFERENCE, AND ANALYTICS

NVIDIA DGX A100

  • 8x NVIDIA A100 Tensor Core GPUs
  • 5 petaFLOPS performance
  • 320 GB of GPU memory
Learn More 
NVIDIA DGX-2

ENTERPRISE AI DEPLOYMENT

NVIDIA DGX-2

  • The first 2 petaFLOPS system that engages 16 fully interconnected GPUs for 10X the deep learning performance
  • Powered by NVIDIA® DGX™ software and an architecture designed for AI-scale built on NVIDIA NVSwitch
  • 6X NVIDIA Tesla V100 GPUs
  • 2 petaFLOPs performance
  • 81,920 NVIDIA CUDA Cores
  • 512 GB of GPU memory
Learn More 

OBTENGA LO MÁS RECIENTE DE NVIDIA SOBRE COMPUTACIÓN DE AI

Productos
Tecnologías
Software
Recursos