BIZON G9000 – 8 NVIDIA Tesla V100 Deep learning and Parallel Computing GPU Server – DGX-1 Alternative – 4-8 Tesla V100 32GB, dual Xeon up to 56 cores

BIZON G9000

8x NVIDIA Tesla V100 Deep Learning Server

Same performance and hardware as DGX-1. 30% less price.

8x NVIDIA Tesla V100 SXM2 (32 GB) with 8-way NVLink hybrid cube-mesh topology. NVIDIA DGX-1 DGX-2 Alternative

Estimated Ship Date: 7-10 Days

Starting at $47,990

System Core
Processors (Intel Xeon Scalable)

Processors (Intel Xeon Scalable)

Memory (DDR4 2666 MHz ECC)

Memory (DDR4 2666 MHz ECC)

Below are recommendations for choosing RAM size.

A Deep Learning computer should have at least as much RAM as GPU memory.

  • 2 x NVIDIA 2080 Ti GPUs: 32 GB (2 x 11 GB of 2080 Ti's memory = 22 GB minimum).
  • 4 x NVIDIA 2080 Ti GPUs: 64 GB (4 x 11 GB of 2080 Ti's memory = 44 GB minimum).
  • If you are working with large images we recommend you 128 GB of memory.
CollapseLearn More
Operating System

Operating System

You can choose multiple operating systems.
  • BIZON Z–Stack (includes Ubuntu 18.04 and latest deep learning frameworks)
    Recommended for Deep Learning. Preinstalled with the Most Powerful Deep Learning Software (including Tensorflow, Torch/PyTorch, Keras, Caffe 2.0 Caffe-nv, Theano, CUDA and cuDNN). We provide BIZON Z-Stack Tool with a user-friendly interface for easy installation and future upgrades. If a new version of any framework is released, you can upgrade with a click of a button and avoid complicated command lines.

  • Windows 10 Pro
    Operating system for wide range of tasks.

CollapseLearn More
Graphics
Graphics Card (SXM2)

Graphics Card (SXM2)

Storage
HDD #1 (Operating System, Applications)

HDD #1 (Operating System, Applications)

HDD #2

HDD #2

Adapters
Network

Network

Service & Support
Warranty/Support

Warranty/Support

The "Life-time Expert Customer Care with 3 / 4 / 5 Year Limited Warranty" includes the following benefits:

  • Life-time access to our expert customer support team for all kinds of computer issues.

  • Three / four / five years (depending on the option you choose) of labor free warranty service (including labor costs for part replacements, and diagnostics).

  • One / two / three years (depending on the option you choose) of direct defective part replacement service directly from BIZON's inventory. After the first / second / third year, BIZON will help replace any defective parts for the customer as long as those parts are still under warranty with the manufacture (most components are covered by 3-5 years warranty from the manufacture).
CollapseLearn More
$47,990
 
In Stock

Need help?

Contact our specialists and get help.

Ask a Question
Use cases
  • Artificial Intelligence
  • Big Data Analytics
  • High-performance Computing
  • Research Lab/National Lab
  • Astrophysics, Business Intelligence
Key Features
  • Dual socket P (LGA 3647) supports Intel Xeon Scalable Processors,3 UPI up to 10.4GT/s
  • Up to 6TB ECC 3DS LRDIMM , up to DDR4-2666MHz ; 24x DIMM slots
  • 4 PCI-E 3.0 x16 (LP) (GPU tray for GPUDirect RDMA), 2 PCI-E 3.0 x16 (LP, CPU tray)
  • Support 8 Tesla V100 with 300GB/s NVLINK
  • 2 10GBase-T LAN with Intel X540
  • 16 Hot-swap 2.5" SAS/SATA drives (Optional 8x NVMe drives supported), 2 NVMe based M.2 SSD
  • 8 Hot-swap 92mm cooling fans
  • 2200W Redundant (2+2) Power Supplies; Titanium Level (96%+)
Ports

Front panel:

  • Power button
  • 2 x RJ45
  • 1 x RJ45 IPMI_dedicated
  • 2 x USB 3.0
  • 1 x VGA
Motherboard

Dual CPU server board.

  • Intel Xeon Scalable Processors (Socket P).
  • Intel C621 chipset.
CPU
Dual Intel Xeon Scalable processors with up to 28 Cores (per CPU).
RAM

  • Up to 768 GB DDR4 2666 MHz ECC Buffered
  • 24 RAM slots.

Note: actual RAM clock speed depends on the CPU specifications and may vary.

Expansion Slots
  • 4 x PCIe 3.0 x16 (LP, GPU tray for GPUDirect RDMA)
  • 2 x PCIe 3.0 x16 (LP, CPU tray) or up to 8x 2.5" NVMe Storage Devices
Storage

Intel C621 Chipset:

  • 8 x SATA 6Gb/s port(s)
  • Support Raid 0, 1, 5, 10
  • 2 x M.2 Socket 3, with M Key, type 2260/2280/22110 storage devices support (PCI-E 3.0 x4)
Graphics Card
Up to 8 Nvidia Tesla V100 32 GB with NVLink hybrid cube-mesh topology.
Onboard LAN
  • Dual 10 Gigabit Ethernet (Intel X540)
  • Dedicated IPMI (Realtek RTL8211E PHY)
  • 100 Gbps Dual Port InfiniBand EDR (Mellanox ConnectX VPI Adapter; QSFP28) (optional)
Chassis
4U Rackmountable

Case Drive Bays:

  • 16 x 2.5" SAS/SATA drive bays
  • 8 x 2.5" NVMe supported (optional)
Cooling System
  • 8 x 92mm RPM Hot-Swap cooling fans
  • 4 x 80mm RPM cooling fans
Power Supply
2000W Redundant (2+2) Power Supplies; Titanium Level (96%+)
Note: G8000 comes with 2 x 2000W Redundant (2+2) power supplies. Total 4 power supplies (2 main + 2 backups).
That means that maximum output power is 4000 W (2 x 2000W) with 4000W (2 x 2000W) for backup power supplies.
If one or both power supplies will fail, the other backup power supply starts running to keep the computer powered up.
The switch between the power supplies is seamless, so as not to interrupt the normal use of the computer.
Power Requirements
100-120V (standard US outlet):
If you have a standard 20 A / 120V US outlet, you can connect one power supply to one dedicated power circuit (each outlet should have its dedicated circuit). In total, you need four (4) 20 A /120V outlets with a dedicated circuit.

Actual power consumption is not the same as maximum power supply power. Power supply power means how much watts you can get from the power supply and your actual power consumption should not be higher than max power supply power. Actual power consumption depends on your configuration (graphics card and CPU).

Power consumption example: G8000 with 8 x Tesla V100 32G B + 2 x Xeon 4116 CPUs = 2400-2600W (100% load for all the GPUs and CPUs).

Power supplies efficiency:
1000W @ 110-120V
2000W @ 220-240V

Information for reference only. Consult your electrician for more details.
Noise

4 x NVIDIA Tesla V100 32 GB:

Idle: 60 db
Max Load (stress test): 70 db

Measured from 6 ft / 1.8 m distance.

Dimensions & Weight

Case

  • Height: 7.0 inches (17.8 cm)
  • Width: 17.6 inches (44.7 cm)
  • Depth: 31.7 inches (80.5 cm)
  • Weight (empty case): 80.0 lbs (36.2 kg)

Shipping Package

  • Height: 48 inches (121.9 cm)
  • Width: 25 inches (63.5 cm)
  • Length: 40 inches (101.6 cm)
  • Weight: 135 lbs (61.2 kg) (shipped on pallet)
Electrical and Operating Requirements
  • Line voltage: 100–240V AC
  • Frequency: 50Hz to 60Hz, single phase
  • Operating temperature: 50° to 95° F (10° to 35° C)
  • Operating Relative Humidity: 8% to 90% (non-condensing)
In the Box
  • Bizon G9000.
  • 4 x power cords (USA).
  • User manual.
  • Benchmark report.
Operating System
  • BIZON Z–Stack (Ubuntu 18.04 (Bionic) with preinstalled deep learning frameworks).
  • Windows Server 2016.
Part Number
bizon-g9000-0119

All specifications are subject to change without notice. The entire materials provided herein are for reference only. Weight varies by configuration and manufacturing process. Advertised performance is based on maximum theoretical interface values from respective Chipset vendors or organization who defined the interface specification. Actual performance may vary by system configuration.

Why Choose BIZON?

Education Pricing and Student Discounts

BIZON offers discounts to academic institutions and students. Contact us for details.

Fast shipping. Fast built times.

Next Day Air US shipping. 3–5 days worldwide shipping.

Money Back Guarantee

Shop with confidence.

Quality You Can Trust

Every product is tested through a rigorous quality assurance process before being shipped to you.

Expert Customers Support

Premiere customer service support with a dedicated tech support team, ready to help you.