Fully Managed Cloud

Get started right away with VESSL's fully managed AWS

Overview

Out-of-the-box, VESSL comes with a fully managed AWS and provides the easiest way to run compute-demanding jobs on the cloud. VESSL Cloud solves the common engineering overheads in ML in the cloud, such as volume mount, runtime customization, spot instances, and model checkpoints — at zero additional cost.

VESSL Cloud provides support for two different 🔗 AWS Regions — Asia Pacific (Seoul, ap-northeast-2) and US West (Oregon, us-west-2). In addition to your default region, you can an additional region on the web.

Users on the Pro plan can take advantage of the powerful GPU instances. Below are the GPU specs and equivalent Amazon EC2 instances that VESSL Cloud supports. Check out the links for more details.

• T4 — 🔗 EC2 G4 Instances

• K80 — 🔗 EC2 P2 Instances

• V100 — 🔗 EC2 P3 Instances

Step-by-step Guide

To add an alternative region follow the following steps.

(1) Click New cluster.

(2) Under VESSL Cloud, select your Region and Cluster.

(3) Click Done to complete the setup and confirm your integration under 🗂️ Clusters.

If you are on the Enterprise plan and wish to deactivate VESSL Cloud for preventive measures, contact [email protected].