WinFast GPU Workstation and Server
Scalable and Optimized System Solution for AI, HPC, AIoT, Big Data and Edge Computing
With market proven GPU and AI expertise, Leadtek provides not only complete solution but also end-to-end certified instructor-led training, workshop, reference architecture and benchmark on NVIDIA GPU-accelerated application to help customer for early adoption.
24-hour online service
Leadtek provides 24/7 online service for customers who purchase system products with pre-installed software. If there are problems related to the use of the product, you are welcome to create a project tracker on the online service system and describe your issue in detail. You can also upload photos as additional explanation. Professional engineers in Leadtek are always available to resolve your problems online.
24-Hour Online Service
Leadtek NVIDIA-Certified System for Enterprise
Systems certified by NVIDIA for accelerated computing
Enterprises today need a scalable computing infrastructure that provides the performance, manageability, security, and scalability to deliver cutting-edge products and services, while increasing their operational efficiencies. NVIDIA-Certified Systems™ enable enterprises to confidently deploy hardware solutions that securely and optimally run their modern accelerated workloads.
NVIDIA-Certified Workstations
Leadtek NVIDIA-Certified desktop workstations are validated for optimum performance and exceptional reliability. They are ideal for running software in areas such as AI, data science, rendering, and visual design.
AI/ML MODEL DEVELOPMENT
ADVANCED VFX
RENDERING
DATA ANALYTICS
ADVANCED VISUALIZATION
PROFESSIONA BROADCASTING
| WS750 | WS850 | WS950 | WS1050 | WS2050 | |
|---|---|---|---|---|---|
| GPU |
|
|
|
|
|
| # of GPU | 1 | 2 | 4 | 2 | 4 |
| CPU | Intel® i7 14700 (8+12)core/2.1GHz x1 | Intel® Xeon® W5-2445/10 Core/3.1GHz x1 | Intel® Xeon® W-3400 processors | Intel® Xeon® 5415+/8Cores/2.9GHz x2 | Intel® Xeon® 5415+/8Cores/2.9GHz x2 |
| RAM | 34GB DDR4-4800 ECC UDIMM x4 | 64GB DDR5-4800 ECC RDIMM x4 | 64GB DDR5-4800 ECC RDIMM x8 | 64GB DDR5-4800 ECC RDIMM x4 | 64GB DDR5-4800 ECC RDIMM x4 |
| Storage |
|
|
|
|
|
| Power supply | 668W Platinum Level Power Supply | 1200W Redundant Power Supplies, Titanium | 2000W Platinum Level Power Supply | 2000W Platinum Level Power Supply | 2000W(1+1) Titanium Level Redundant Power Supplies |
| Learn More | Learn More | Learn More | Learn More | Learn More |
NVIDIA-CERTIFIED SERVERS
NVIDIA-Certified servers create the essential platform for the evolution of enterprise data centers, delivering infrastructure that can handle a diverse range of accelerated workloads. The certification test suite exercises the performance and functionality of a configured server by running a set of software that represents a wide range of real-world applications.
COMPUTE
AI training and inferencing, data analytics, HPC
GENERAL PURPOSE
Visualization, rendering, AI, virtual workstations
HIGH-DENSITY VIRTUALIZATION
Virtual desktops, Virtual Workstations
ENTERPRISE EDGE
Edge inferencing in controlled environments
INDUSTRIAL EDGE
Edge inferencing in industrial or rugged environments| Model | WinFast GS2040T | WinFast GS2045T | WinFast GS2050T | WinFast GS4840 | WinFast GS4850T | WinFast GS4855ST |
|---|---|---|---|---|---|---|
| GPU | A6000/A100 | A100 | RTX 6000 Ada | A6000/A100 | RTX 6000 Ada | RTX 6000 Ada |
| # of GPU | 4/6 | 4 | 4 | 8 | 8 | 8 |
| CPU | Intel® Xeon® Gold 5320, 24 Cores/2.2GHz x2 | AMD EPYC™ 7402, 24 Cores/2.8GHz x 2 | Intel® Xeon® Gold 5418Y, 24 Cores/2.0GHz x2 | Intel® Xeon® Gold 5320, 24 Cores/2.2GHz x2 | Intel® Xeon® Gold 5418Y, 24 Cores/2.0GHz x2 | AMD® Genoa 9254 24Cores/2.4GHz x2 |
| RAM | 64GB DDR4-3200 RDIMM x8 | 64GB DDR4-3200 RDIMM x8 | 64GB DDR4-4800 RDIMM x8 | 64GB DDR4-3200 RDIMM x16 | 64GB DDR4-4800 RDIMM x16 | 64GB DDR4-4800 RDIMM x16 |
| Storage |
|
|
|
|
|
|
| Power supply | 2600W (1+1) Titanium Level Redundant Power Supplies | 2200W (1+1) Platinum Level Redundant Power Supplies | 2600W (1+1) Titanium Level Redundant Power Supplies | 2200W (2+2) Titanium Level Redundant Power Supplies | 2700W (2+2) Titanium Level Redundant Power Supplies | 2000W (2+2) Titanium Level Redundant Power Supplies |
| Network Adapter | MELLANOX ConncetX-6Dx x 2 | |||||
| Ethernet/Infiniband Switch | SN2100- 100 Gigabit Ethernet Switch with 16 QSFP28 ports Plus cables: Mellanox passive copper for 100 Gigabit Ethernet | |||||
| Learn More | Learn More | Learn More | Learn More | Learn More | Learn More | |
NVIDIA-Powered
Data Science and Deep Learning Workstation
for AI Development
WinFast RTX AI Workstation
Leadtek provides WinFast RTX AI Workstation, a complete line of NVIDIA-powered data science workstations featuring Quadro® RTX™ GPUs and leading class data science software. Maximize productivity and lower the cost of your AI development projects. Highly scalable and pre-configured complete system enable time to market, increase efficiency in the organization and to unlock the value from their combined data sets.
Learn More
Delivering Data Science Value
Maximize Productivity
Take advantage of a fully integrated hardware and software stack that is ready out-of-the-box for instant productivity gains.
Faster Model Development and Training
Iterate and move to production clusters faster with high performance Quadro RTX GPUs.
Framework Diagnostic and System Recovery
Simple and intuitive tool for framework diagnosic and system recovery.
Enterprise-Class
The workstation is backed with 3 year warranty for hardware components and the software stack is optimized and tested for world-class support so researcher and developer can rely on
Integrated with GPU-Accelerated Software
NVIDIA-powered data science workstations come with a comprehensive stack of tested and optimized data science software built on NVIDIA CUDA-X AI. a collection of over 15 libraries that enable modern computing applications to benefit from NVIDIA’s Quadro RTX GPU accelerated computing platform. This stack features RAPIDS data processing and machine learning libraries, NVIDIA optimized TensorFlow, PyTorch, Caffe and other leading data science software, providing enterprises with accelerated workflows for faster data preparation and model training.
Core Components and Frameworks
| WinFast RTX AI Workstations | WS750 | WS850 | WS950 | WS1050 | WS2050 |
|---|---|---|---|---|---|
| GPU |
|
|
|
|
|
| # of GPU | 1 | 2 | 4 | 2 | 4 |
| CPU | Intel® i7-14700 (8+12)core/2.1GHz x1 | Intel® Xeon W5-2445/10 Core/3.1GHz x1 | Intel® Xeon W5-3435X/ 16Core/3.1GHz x1 | Intel® Xeon 5415+/8Cores/2.9GHz x2 | Intel® Xeon 5415+/8Cores/2.9GHz x2 |
| RAM | 32GB DDR5-4800 ECC UDIMM x4 | 64GB DDR5-4800 ECC RDIMM x4 | 64GB DDR5-4800 ECC RDIMM x8 | 64GB DDR5-4800 ECC RDIMM x4 | 64GB DDR5-4800 ECC RDIMM x8 |
| Storage |
|
|
|
|
|
| Power Supply | 668W Platinum Level Power Supply | 1200W Platinum Level Power Supply | 2000W Platinum Level Power Supply | 2000W Platinum Level Power Supply | 2000W(1+1) Titanium Level Redundant Power Supplies |
| Learn More | Learn More | Learn More | Learn More | Learn More |
AI/Deep Learning System Solution
The GPU-Optimized Servers for DL Workloads to Accelerate Your AI Journey
Implementing the right AI/deep learning solution can be complex. Leadtek provides versatile server, accelerator, DL software and technical consult as end-to-end solution. Our proven, practical approach, validated solutions and partners, GPU-optimized infrastructure and AI software platform – reduce complexity and help you realize the value of your data faster helping you gain a competitive edge.
Scalable Deep Learning
Offering a wide range of customizable options from desktop to data center to meet your budget and project need.
Pre-Configured System
Leadtek AI/DL system solutions are fully turnkey; GPU-accelerated comprehensive software stack of data science software built on NVIDIA CUDA-X AI with RAPIDS, ML libraries and frameworks. Therefore, you avoid the time-consuming of configuration and setup.
Suggested Configurations
| WinFast RTX AI Server | WinFast GS2040T | WinFast GS2045T | WinFast GS2050T | WinFast GS4840 | WinFast GS4850T | WinFast GS4855ST |
|---|---|---|---|---|---|---|
| GPU | A6000/A100 | A100 | RTX 6000 Ada | A6000/A100 | RTX 6000 Ada | RTX 6000 Ada |
| # of GPU | 4/6 | 4 | 4 | 8 | 8 | 8 |
| CPU | Intel® Xeon® Gold 5320, 24 Cores/2.2GHz x2 | AMD EPYC™ 7402, 24 Cores/2.8GHz x2 | Intel® Xeon® Gold 5418Y, 24 Cores/2.0GHz x2 | Intel® Xeon® Gold 5320, 24 Cores/2.2GHz x2 | Intel® Xeon® Gold 5418Y, 24 Cores/2.0GHz x2 | AMD® Genoa 9254 24Cores/2.4GHz x2 |
| RAM | 64GB DDR4-3200 RDIMM x8 | 64GB DDR4-3200 RDIMM x8 | 64GB DDR4-3200 RDIMM x8 | 64GB DDR4-3200 RDIMM x16 | 64GB DDR5-4800 RDIMM x16 | 64GB DDR5-4800 RDIMM x16 |
| Storage |
|
|
|
|
|
|
| Power Supply | 2600W (1+1) Redundant Power Supplies, Titanium | 2200W (1+1) Redundant Power Supplies, Platinum | 2600W (1+1) Redundant Power Supplies, Titanium | 2200W (2+2) Redundant Power Supplies, Titanium | 2700W (2+2) Redundant Power Supplies, Titanium | 2000W (2+2) Redundant Power Supplies, Titanium |
| Learn More | Learn More | Learn More | Learn More | Learn More | Learn More |
How GPU Speeds Up AI &Deep Learning Computing
Visit Leadtek AIHigh Performance Computing Solution
Accelerate to Discovery and Solve Your Most Difficult Computing Challenges
Accelerate your most demanding HPC and hyperscale data center workloads with WinFast GPU Servers. Researchers and data scientists can now parse petabytes of data orders of magnitude faster than they could using traditional CPUs, in applications ranging from energy exploration to scientific research. The high performance server come in a variety of HPC configurations, can be integrated with leading parallel computing technologies NVIDIA GPUs, and deliver the breakthrough compute power necessary for critical innovation and problem solving in this fast-paced Information Age.
Suggested Configurations
| WS945 | |
|---|---|
| Form Factor | Tower/5U Rackable |
| CPU | 1x AMD Ryzen™ Threadripper™ PRO/16 Cores |
| RAM | 256GB |
| GPU | 4x NVIDIA A6000/A5000/RTX6000/RTX8000 |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 2x 1.92TB SATA HDD |
| Learn More |
| WS2040 | |
|---|---|
| Form Factor | Tower/4U Rackable |
| CPU | 2x Xeon Gold/12 Cores |
| RAM | 256GB |
| GPU | 4x NVIDIA A100/A40/A30/A10/T4/ A6000/A5000/RTX6000/RTX8000 |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 2x 1.92TB SATA HDD |
| Learn More |
| GS2045T | |
|---|---|
| Form Factor | 2U |
| CPU | 2x AMD EPYC/24 Cores |
| RAM | 384GB |
| GPU | 4x NVIDIA A100/A40/A30/A10/T4/ A6000/A5000/RTX6000/RTX8000 |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 2x 1.92TB SATA HDD |
| Learn More |
| GS4840 | |
|---|---|
| Form Factor | 4U |
| CPU | 2x Xeon Gold/28 Cores |
| RAM | 768GB |
| GPU | 10x NVIDIA A100/A40/A30/A10/T4/ A6000/A5000/RTX6000/RTX8000 |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 2x 1.92TB SATA HDD |
| Learn More |
| GS4845 | |
|---|---|
| Form Factor | 4U |
| CPU | 2x AMD EPYC/32 Cores |
| RAM | 768GB |
| GPU | 8x NVIDIA A100/A40/A30/A10/T4/ A6000/A5000/RTX6000/RTX8000 |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 2x 1.92TB SATA HDD |
| Learn More |
The Most Powerful Virtual Workstation
Professionals To Do Their Work From Anywhere
Grounded in the NVIDIA EGX platform, the EGX server for professional visualization is configured with NVIDIA vGPU software to create powerful virtual workstations that enable professionals to do their work from anywhere while increasing productivity, improving data center utilization, and reducing IT costs and maintenance. Provision multiple, easy-to-manage virtual workstations or a combination of virtual workstations and render nodes from a single EGX server with NVIDIA RTX Virtual Data Center Workstation (NVIDIA RTX)
Certified by NVIDIA EGX Server for Virtual Workstation
| GS2030T | |
|---|---|
| Form Factor | 2U |
| CPU | 2x Xeon Gold/24Cores |
| RAM | 960GB or above |
| GPU | 5x Quadro RTX6000 Passive/RTX8000 Passive |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 2x 4TB SATA |
| Learn More |
Scalable Visualization Solutions
To Easily Deploy and Manage Large-display
Drive dynamic, large-scale visualization with EGX server configured for scalable visualization, which allows you to easily deploy and manage large-display environments from the data center. Whether you’re developing the next theme park attraction or creating multi-sided cave automatic virtual environments (CAVE), you can now drive the most immersive visual experiences.
Certified by NVIDIA EGX Server for Scalable Visualization
| GS4830AT | |
|---|---|
| Form Factor | 4U |
| CPU | 2x Xeon Gold/16Cores |
| RAM | 256GB or above |
| GPU |
|
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 2x 1.92TB SATA HDD |
| Learn More |
Rendering Solution
High-Performance, Flexible Rendering from Desktop to the Data Center
Render final frames or boost local workstation performance with the power of GPU acceleration. By utilizing EGX server with RTX™-accelerated software designers and artists can render jobs offline and produce final ray-traced images at a fraction of the time of traditional CPU-based render nodes. Even achieve fully interactive, photorealistic visualization in the application viewport by connecting to one or more EGX server to boost your desktop performance.
Certified by NVIDIA EGX Server for Rendering
| WS2030 | |
|---|---|
| Form Factor | Tower/4U |
| CPU | 2x Xeon Gold/12 Cores |
| RAM | 256GB or above |
| GPU | 4x Quadro RTX6000 (Passive ) RTX8000 (Passive) |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 1x 1.92TB SATA HDD |
| Learn More |
| GS1030ST | |
|---|---|
| Form Factor | 1U |
| CPU | 1x Xeon Gold/16Cores |
| RAM | 128GBor above |
| GPU | 2x Quadro RTX6000 (Passive ) RTX8000 (Passive) |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 1x 1.92TB SATA HDD |
| Learn More |
| GS4830AT | |
|---|---|
| Form Factor | 4U |
| CPU | 2x Xeon Gold/12Cores |
| RAM | 512GB or above |
| GPU | 8x Quadro RTX6000 (Passive ) RTX8000 (Passive) |
| System Drive | 1x 960GB NVMe PCI-E |
| Data Drive | 1x 1.92TB SATA HDD |
| Learn More |
Workstation
Learn more
Server
Learn moreWinFast Workstation
Extreme Performance for Modeling, Animation and AI/Deep Learning
Overview
WinFast Workstation is designed for technical and creative professionals who demand the highest productivity and reliability for the heaviest of workload. Leadtek best-of-class workstations is optimized and configured with NVIDIA GPU for the workload ranging from entry level workload to multi-tasking in complex production pipeline.
Technology
- Highly optimized with NVIDIA GPU for performance and reliability
- Balanced multi-GPU thermal and acoustic design
- Flexible configuration and GPU for versatile workload demands
Application
- AI/Deep Learning Development
- 2D Image Processing
- CAD Design
- Engineering Simulation
- 3D Animation and Modeling
- Rendering
- VR
Key Features
Form Factor
Slim and Mid Tower
Processor
Single or dual Intel® Xeon® Scalable processors up to 205W TDP
GPU
Up to 4 NVIDIA GPUs with NVLink Technology
Networking
Up to dual 10GbT LAN with IPMI 2.0
System Memory
Up to 16 DIMM slot and 4TB DDR4 memory
Storage Drive
Up to 8 drives with 2.5" and 3.5" SAS3/SATA3 options (4 NVMe optional), 1 M.2 NVMe drive
Power Supply
Up to Redundant Titanium Level Power Supplies
Warranty
3-Year hardware services
Workstation Line
1 CPU + 1 GPU
Entry
- PLM
- CAD
- Basic DCC
- Medical Imaging
- Entry DL/ML
1 CPU + 2 GPU
Mainstream
- Mid-size CAD
- Advanced DCC
- Simulation
- Medical Imaging
- DL/ML
1 CPU + 4 GPU
High End
- Complex CAD
- Complex DCC
- Rendering
- 3D Medical Imaging
- Simulation
- Seismic Exploration
- Large Scale DL/ML
2 CPU + 4 GPU
Supreme
- Complex CAD
- Photo-realistic Rendering
- 3D Medical Imaging
- Simulation
- Seismic Exploration
- Large Scale DL/ML
WinFast GPU Server
Optimized and GPU-accelerated for AI/Deep Learning and HPC
Overview
Built on our years of experience in the server industry, Leadtek introduces a comprehensive range of GPU-centric WinFast GS server solutions - designed to deliver, performance, scalability and reliability in the applications such as AI/Deep Learning, High Performance Compute (HPC), virtual graphics, render farm and other GPU-accelerated intensive workloads.
Technology
- Highly Optimized GPU-to-GPU Interconnect Architectures for best efficiency and low latency
- Optimized multi-GPU thermal designs for highest performance and reliability
- Leading NVIDIA® GPU architectures HGX-2 platform with NVLink™ and NVSwitch™
Application
- AI/Deep Learning
- High Performance Computing
- Scientific Research
- Data Analytics
- 3D Animation and Modeling
- 3D Graphics Accelerated Virtualization
- Rendering
Form Factor
1U, 2U, 4U and 10U Rackmount
Processor
Single or dual Intel® Xeon® Scalable processors up to 205W TDP
NVIDIA GPU
Up to 16 GPUs in 10U
Networking
Up to dual 10GbT LAN with optional IB FDR
System Memory
Up to 24 DIMM slot and 6TB DDR4 memory
Storage Drive
Up to 16 NVMe drives with 2.5" and SAS3/SATA3 options, Up to 2 M.2 NVMe drives
Power Supply
Redundant Platinum/Titanium (96%+) Level PSU
Warranty
3-Year hardware services
Server Line
1U / 2 GPU
1U Server
- Virtual GPU (GRID)
- Remote Workstation
- Small Scale Render Farm
2U / 6 GPU
2U Server
- Virtual GPU (GRID)
- Render Farm
- DL/ML
- High Performance
- Computing
- 3D Medical Imaging
4U / 8 GPU
4U Server
- Cloud Computing
- Large Scale DL/ML
- High Performance
- Computing
4U / 8 GPU
4U Server
- Cloud Computing
- Large Scale DL/ML
- Render Farm
- High Performance
- Computing
4U / 10 GPU
4U Server
- Cloud Computing
- Large Scale DL/ML
- High Performance
- Computing
10U / 16 SXM3 GPU
4U Server
- Hyper Scale Cloud
- Hyper Scale DL/ML
- High Performance
- Computing
Leadtek AI Forum
Realizing Your Future of AI
Visit Leadtek AI Forum
What We Do
Leadtek AI Forum is an online discussion community where accessible to Leadtek’s customer or registered developer discussing their queries related to Leadtek’s AI solutions or specific ML/DL technologies to help individual AI development and deployment.
Why Join
Whether you’re looking for ML/DL solutions for your AI project, the Leadtek AI Forum is the place to be. With the entire ecosystem including software and hardware, discuss ML/DL technological evolution, trend, or solution with Leadtek experienced AI experts, NVIDIA DLI certified instructors to advance your AI investment.
© 2024 Leadtek Research Inc. All Rights Reserved