Servidor de IA Lenovo ThinkSystem SR680a V3 | 8x NVIDIA HGX H200 | Doble Intel Xeon 8568Y+ | TruDDR5 de 2 TB
  • Categorías de productos:Servidores
  • Número de parte:Lenovo ThinkSystem SR680a V3
  • Disponibilidad:In Stock
  • Condición:Nuevo
  • Características del producto:Listo para enviar
  • Pedido mínimo:1 unidad
  • Precio de lista era:$439,999.00
  • Su Precio: $410,000.00 Ahorra $29,999.00
  • Chatear Ahora Enviar Email

Respire tranquilo. Devoluciones aceptadas.

Envío: El envío internacional de artículos puede estar sujeto a procesamiento aduanero y cargos adicionales. Ver detalles

Entrega: Por favor, permita tiempo adicional si la entrega internacional está sujeta a procesamiento aduanero. Ver detalles

Devoluciones: Devoluciones aceptadas durante 14 días. El vendedor paga el envío de devolución. Ver detalles

Envío GRATIS. Aceptamos órdenes de compra NET 30 Días. Obtenga una decisión en segundos, sin afectar su crédito.

Si necesita una gran cantidad del producto Lenovo ThinkSystem SR680a V3, llámenos a nuestro número gratuito Whatsapp: (+86) 178-3212-0331 o solicite un presupuesto en el chat en vivo y nuestro gerente de ventas se comunicará con usted en breve.

Lenovo ThinkSystem SR680a V3 AI Server | 8x NVIDIA HGX H200 | Dual Intel Xeon 8568Y+ | 2TB TruDDR5

Keywords

lenovo server, ai server, gpu server, buy server hardware, enterprise servers, server hardware, server parts

Description

Discover the pinnacle of artificial intelligence infrastructure with the Lenovo ThinkSystem SR680a V3. Built on the robust ThinkSystem SR680a V3 H200 GPU Base, this ai server is engineered for the most demanding workloads. The centerpiece of this system is the ThinkSystem NVIDIA HGX H200 141GB 700W 8-GPU Board, providing unprecedented acceleration for complex computational environments.

Driving this elite gpu server are two Intel Xeon Platinum 8568Y+ processors. Each processor delivers 48 cores, operates at 2.3GHz, and features a 350W power footprint. To ensure the processors and GPUs are continuously fed with data without bottlenecks, the system is populated with 32 units of 64GB TruDDR5 5600MHz (2Rx4) RDIMMs.

The storage architecture of these enterprise servers is designed for blistering speed and reliability. It features eight 2.5" U.2 PM9A3 3.84TB Read Intensive NVMe PCIe 4.0 x4 HS SSDs for massive data payloads. The operating system is securely supported by two M.2 PM9A3 960GB Read Intensive NVMe PCIe 4.0 x4 NHS SSDs. These M.2 drives are managed by Intel VROC (VMD NVMe RAID) Standard and are configured in an M.2 NVMe Array 1 RAID 1 setup.

High-bandwidth network connectivity is paramount when you buy server hardware for scale-out clusters. This configuration is exceptionally equipped with eight NVIDIA ConnectX-7 NDR OSFP400 1-Port PCIe Gen5 x16 InfiniBand Adapters. Additionally, it includes an NVIDIA BlueField-3 B3220 VPI QSFP112 2P 200G PCIe Gen5 x16 Adapter. Power delivery is safeguarded by eight 2600W 230V Titanium Hot-Swap Gen2 Power Supply v4 units, arranged for N+N Redundancy With Over-Subscription.

Key Features

  • Elite GPU Base: ThinkSystem SR680a V3 H200 GPU Base.
  • Ultimate Accelerator: ThinkSystem NVIDIA HGX H200 141GB 700W 8-GPU Board.
  • High-Core Compute: Dual Intel Xeon Platinum 8568Y+ 48C 350W 2.3GHz Processors.
  • Massive InfiniBand Networking: 8x NVIDIA ConnectX-7 NDR OSFP400 1-Port PCIe Gen5 x16 InfiniBand Adapters.
  • Advanced DPU: ThinkSystem NVIDIA BlueField-3 B3220 VPI QSFP112 2P 200G PCIe Gen5 x16 Adapter.
  • Titanium Efficiency Power: 8x ThinkSystem 2600W 230V Titanium Hot-Swap Gen2 Power Supply v4.
  • Enhanced Security: Includes TPM 2.0 with Secure Boot feature.

Configuration

Component Description / Part Number Qty
System Base ThinkSystem SR680a V3 H200 GPU Base (C1EL) 1
Processor (CPU) Intel Xeon Platinum 8568Y+ 48C 350W 2.3GHz Processor (BYWF) 2
GPU Board ThinkSystem NVIDIA HGX H200 141GB 700W 8-GPU Board (C1HM) 1
Memory (RAM) ThinkSystem 64GB TruDDR5 5600MHz (2Rx4) 10x4 RDIMM (C5H9) 32
Data Storage ThinkSystem 2.5" U.2 PM9A3 3.84TB Read Intensive NVMe PCIe 4.0 x4 HS SSD (BXM9) 8
Boot Storage ThinkSystem M.2 PM9A3 960GB Read Intensive NVMe PCIe 4.0 x4 NHS SSD (BXMH) 2
Backend Network (GPU) ThinkSystem NVIDIA ConnectX-7 NDR OSFP400 1-Port PCIe Gen5 x16 InfiniBand Adapter (BQ1N) 8
Frontend Network (DPU) ThinkSystem NVIDIA BlueField-3 B3220 VPI QSFP112 2P 200G PCIe Gen5 x16 Adapter (BVBG) 1
Power Supplies ThinkSystem 2600W 230V Titanium Hot-Swap Gen2 Power Supply v4 (C4HK) 8

Compatibility

The ThinkSystem SR680a V3 H200 GPU Base is specifically designed to house the ThinkSystem NVIDIA HGX H200 141GB 700W 8-GPU Board. To operate the eight ThinkSystem 2600W 230V Titanium power supplies at peak efficiency, the server infrastructure relies on High voltage (200V+) input. Additionally, the dual M.2 boot storage solution requires the Intel VROC (VMD NVMe RAID) Standard for M.2 to run its RAID 1 array safely.

Usage Scenarios

This lenovo server is uniquely architected for massive ai server workloads. The extreme computational power provided by the ThinkSystem NVIDIA HGX H200 8-GPU Board, which draws 700W per GPU, is essential for deploying large-scale neural networks and advanced deep learning models.

For operations requiring immense data ingestion, the server hardware provides exceptional throughput via the eight ThinkSystem NVIDIA ConnectX-7 NDR OSFP400 InfiniBand Adapters. This ensures that when clustered with other enterprise servers, the GPUs communicate directly without latency bottlenecks.

System administrators utilizing server parts in dense environments can leverage the Front Operator Panel with Quad Line LCD Display for immediate physical monitoring. Furthermore, the system is tuned with the "Efficiency - Favoring Performance Mode" operating mode selection, ensuring it balances maximum compute output with optimized power consumption.

Frequently Asked Questions

Q: What type of RAID does the M.2 boot drive use?
A: The system uses Intel VROC (VMD NVMe RAID) Standard for M.2, which is configured specifically as an M.2 NVMe Array 1 in a RAID 1 mirror.

Q: How much memory is included in this server configuration?
A: This server includes 32 modules of ThinkSystem 64GB TruDDR5 5600MHz (2Rx4) RDIMMs.

Q: What are the power supply specifications for this massive GPU system?
A: The server is equipped with eight ThinkSystem 2600W 230V Titanium Hot-Swap Gen2 Power Supply v4 units. They are expertly configured for N+N Redundancy With Over-Subscription.

Q: Does the system include a Data Processing Unit (DPU) for network offloading?
A: Yes, it features one ThinkSystem NVIDIA BlueField-3 B3220 VPI QSFP112 2P 200G PCIe Gen5 x16 Adapter.

PRODUCTOS RELACIONADOS CON ESTE ÍTEM
XFusion  G5500V7 /  8468*2 / 32G  5600 *4 /1.92T *3 / NVDIA  RTX4090D-48G*8 / 9560-4G / 3000W*4 Recommend
Servidor en rack de misión crítica Dell PowerEdge R770 con doble Intel Xeon 6530P, memoria DDR5 de 1,5 TB, RAID empresarial, redes de 25 GbE e iDRAC10 Recommend
Servidor en rack de misión crítica Dell PowerEdge R770 con Intel Xeon 6530P dual, DDR5 de 768 GB, RAID empresarial y redes de 25 GbE Recommend
Servidor GPU AI xFusion G5500 V6 con doble Intel Xeon Gold 6330, 8× NVIDIA A800 80GB SXM4, 200Gb InfiniBand y almacenamiento NVMe Recommend
Servidor en rack HPE ProLiant DL380 Gen11 2U con doble Intel Xeon Gold 6430, 256 GB DDR5, doble 25 GbE, 32 Gb FC, RAID MR408i-o y garantía de 3 años Recommend
Dell PowerEdge R760 / 5418Y*2/ 64 DDR5 *12 /20TSATA*12/ H755/ 1400*2 Recommend
Solución de servidor GPU con IA de alta densidad: xFusion FusionServer G5500 V6 con 8× NVIDIA A800 80 GB SXM4, doble Intel Xeon Gold 6330 y almacenamiento NVMe Recommend
Inspur NF5468M7 (8*2.5SFF)6530*2/ 16*64TB DDR5 ECC / 8× NVIDIA L40S 48GB /2× 1.92TB SSD RAID1 / 4× 7.68TB NVMe Recommend