Cloud vGPU

Rent a virtual server with GPU
NVIDIA® Tesla® GPU Cloud server for tasks that demand high powerful processing power.

Try for free

vGPU

Cloud4Y offers virtual servers with NVIDIA® GPUs that take deep learning, machine learning, high-performance computing (HPC) and graphics rendering applications to new levels of performance.

Depending on the tasks, we offer GPU- accelerated cloud servers with NVIDIA® Tesla® P100, M40 and M60 GPUs.

Rent a render farm based on Tesla M40 graphics accelerators.

Service
Workstation GPU
Cost
from 3 000 ₽ per month

GPU Cloud computing


Deep Learning / Machine Learning

Use the world's most powerful accelerators for in-depth training of NVIDIA® Tesla® neural networks.

The virtual server with GPU is optimized for 24/7 operation and can train neural networks up to 8 times faster than the CPU.

High Performance Computing (HPC)

Perform tasks that require serious computing power faster and more accurately. The Tesla® M60 and Tesla® M40 Accelerators provide the performance you need to handle big data.

NVIDIA® Tesla® P100 GPUs guarantee 5% higher accuracy of computations using double precision (compared to single precision) while speedup is 20 times higher in double precision mode than popular NVidia® GTX 1080 models.


GPU Accelerated Virtual Desktop Infrastructure

NVIDIA GRID and VDI technologies (VMware Horizon View solution) allow you to work remotely with highly loaded applications in the cloud.

GPU Cloud Servers for graphics and video

NVIDIA® Tesla® graphics cards with super fast storage provide excellent performance for the most demanding 3D graphics, video and rendering applications - AutoCAD, Sony Vegas Rgo, 3DS Max, Autodesk Maya, Adobe products and more.





Why trust Cloud4Y
11 years in cloud computing
Since 2009 the company has been successfully operating in the Russian and international cloud services markets.
Reliable infrastructure
4 TIER III data centers, Enterprise level hardware and software: HP, Cisco, Juniper, NetApp, VMware, Veeam, Microsoft, etc.
SLA 99.982%
Optical ring, MetroCluster and redundancy mechanisms guarantee fault tolerance of services up to SLA 99.99%.
Transparent Billing Options
Hourly billing and pay-as-you-go allow you to pay only for the resources consumed.
Geo-Distributed Backup
Automatic backup (14 restore points) in a separate remote data center.
Flexible scalability
You can add and reduce the amount of resources without need to contact technical support service.
24/7 technical support
If any technical issue occur, our team of support experts is available round-the-clock, response time is 10 minutes.
Partner Program
Earn up to 35% of your annual contract. White Label is available.

GPU Cloud Benefits

  • NVIDIA® Tesla® P100 - the world's most powerful GPUs
  • x10 increase in computing speed (compared to CPU)
  • Savings capital costs up to 70%
  • Payment for actually consumed resources
  • Access to graphic resources from anywhere in the world
  • Data protection on УЗ1-4, 1К, 1Г


FAQ

Frequently asked questions (FAQ)

1. How does it work?
The Cloud GPU service is delivered via VDI that uses computing resources such as the CPU, RAM and cloud graphics (vGPU). You connect to this desktop from anywhere in the world via the Internet.

2. What software is used to provide this service?
VMware Horizon View is used to connect to the desktop. The Window Server 2016 operating system will be installed inside the desktop itself.

3. Do I need a powerful PC?
The performance of your PC is almost irrelevant. The possibility to install a desktop access program (VMware Horizon View) on your PC is only thing needed.

4. What kind of communication channel is needed?
For normal operation, a stable Internet channel of 10 Mb/s and above with the lowest possible latency is required. With high latency values in the communication channel, there will be a delay in the response to your actions with the mouse, keyboard, etc. and lags in dynamically changing images on PC screen. Where does high Latency value come from: traffic from your PC to VDI through a large number of intermediate nodes, poor communication channels of your ISP and/or between intermediate nodes, large physical distance of your PC from our data center, etc.

5. What applications can I use on my remote desktop with GPU?
You can install any software, including compute-intensive applications, such as Sony Vegas Pro, GRAPHISOFT, Solidworks, Compass, Nanocad, T-Flex, Polymatics, Adobe CC (in particular, Adobe After Effects, Adobe Photoshop, Adobe Illustrator, Adobe Premiere, etc.), Autodesk (AutoCAD 2017, AutoCAD Electrical 2017, AutoCAD Map 3D 2017, 3DS Max, Revit 2017, Inventor 2017, Vault 2017).

6. Do you offer GPU rental on a per hour basis?
No. We provide this service for at least 1 month.

7. How much does it cost?
On our web-site you can see the cost of ready-made solutions with a strictly defined configuration. But you can configure individual server with as many CPU, RAM, vGPU and disk space you need to meet your requirements.

8. Can I change the server parameters?
Yes, you can. If you need more gigabytes of vGPU or processor cores while using, you have the option of increasing these settings in just a few minutes.

9. Is it possible to use your GPUs for mining?
Yes, but we do not recommend it. The revenue from crypto-mining is uncertain, while you have to pay for our services in complete.

More answers to your questions can be found in the knowledge base. If you do not find an answer to your question - post it to our consultant online using online-chat or send an inquiry using the support ticket system.

Send a request
Let our managers know if you are interested in a solution or a service. They will contacts you within 2 hours.
You also can request a free trial access here