About Cloud vGPU
Take your AI/ML applications to the next level by training and running your models on OWS Cloud vGPU with NVIDIA H100 GPUs.
Get dependable uptime with our 99.99% SLA, simple security tools, and predictable pricing with OWS's Elastic Compute Service, called Planets.
Sensational AI performance
Powered by a new NVIDIA transformer engine and fourth-gen Tensor cores, H100 delivers up to 9x faster AI training and up to 30x faster AI inference*.
Scale your models, not your costs
Combine eight H100 GPU nodes with up to 3.2 TBps GPU interconnect to train and run your most complex models.
Build with speed
Whether it is for large-language models, object detectors, foundation models or any GPU-intensive models, OWS Cloud vGPU makes it easy to spin up a GPU instance in seconds.
Global edge locations
We operate every aspect of our points of presence, so you have a single partner for your global presence.
24x7 support
We’re to help with questions and implementation tips. Contact our support specialists any time of day.
15-second deploys
Deploy servers with the most popular Operating Systems in 15 seconds. All OSs that can't be deployed instantly are deployed in just 10 minutes.
Single-tenant servers
Deploy single-tenant servers for more performance, control, and no risk of noisy neighbors.
1 GB traffic bandwidth per hour per server
Each server gets 1 GB of free traffic per hour, which is automatically added to your hourly traffic bandwidth quota.
Shared traffic pool
Servers in the same region share traffic bandwidth quotas. This means you don't have to worry about separate servers, and you'll have one place to manage everything related to your traffic.
FAQ
Questions About Planets
What are OWS Planets?
OWS Planets are Linux-based virtual machines (VMs) that run on top of virtualized hardware. Each Planet you create is a new server you can use, either standalone or as part of a larger, cloud-based infrastructure.
Do you provide SLA for Planets?
OWS provides a 99.99% uptime SLA for both Planets and Volumes Block Storage.
Which Planets size should I choose?
Choosing the right Planets plan depends on your workload. Please refer to this article that explains the differences between shared and dedicated vCPUs, and goes into detail on each Planets plan.
What regions are the Planets in?
Planets are available in all regions. Most other Planet types are available in all regions. Learn more about regional availability in our Product Docs.
Can I resize my Planets?
Yes. Resizing a Planet, also known as vertical scaling, increases the amount of resources a Planet has. There are two resizing options for Planets:
(1) CPU and RAM only. This option lets you increase or decrease the amount of CPU and RAM available to a Planet.
(2) OS Disk. This option permanently increases the size of a Planet's os disk.
Will I be billed if I power off my Planets?
Yes, when you power off your Planet, you are still billed for it. This is because your disk space, CPU, RAM, and IP address are all reserved, even while powered off. Therefore, charges are made until you destroy the instance.