Getting Started
Get up and running with your first virtual machine
Get up and running with your first virtual machine in just a few steps.
First Steps
Recommended setup process:
Create an Account
Create an account using email/password or Google OAuth for quick access to the platform.
Sign up at: barrack.ai/signup
Complete Your Profile
Navigate to My Account section and complete your profile with billing information (required for payments).
Required information:
- Full name
- Complete address
- Postal code
- Country (determines currency)
Add Credits
Purchase credits to pay for your virtual machines and storage.
Minimum amounts:
- USD: $1.00
- EUR: €1.00
- INR: ₹100.00
Deploy Your VM
Use the Deploy VM page to create your SSH key and launch your virtual machine.
Authentication & Access
Multiple ways to access your account:
Email & Password
Traditional login with your registered email and password
- Secure authentication
- Password recovery available
- Login at barrack.ai/login
OTP Login
Secure one-time password login via email
- Enhanced security
- No password required
- Email verification
Google OAuth
Quick sign-in with your Google account
- Single sign-on
- No additional password
- Instant access
Password Recovery
Email-based password reset with secure verification
- Secure recovery process
- Email verification
- New password setup
Prerequisites Checklist
Before deploying your first VM, ensure you have:
- Complete Profile - Billing information must be filled out
- Valid Address - Street address, city, postal code
- Country Selection - Determines your billing currency
- Contact Information - Phone number (optional but recommended)
- Complete Profile - Billing information must be filled out
- Valid Address - Street address, city, postal code
- Country Selection - Determines your billing currency
- Contact Information - Phone number (optional but recommended)
- Sufficient Credits - Minimum balance required:
- USD: Minimum $1.00
- EUR: Minimum €1.00
- INR: Minimum ₹100.00
- Payment Method - Valid for purchasing additional credits
- SSH Key - Created during VM deployment at barrack.ai/deploy
- Public IP - Enabled during or after deployment
- Key Download - Private key saved to Downloads folder
- Region Selection - Choose the region closest to your users
- GPU Requirements - Determine which GPU type you need
- Storage Needs - Plan for additional storage volumes if needed
Your profile must be complete with billing information (address, postal code, country) before you can create volumes or make purchases.
SSH Connection Guide
Step 1: SSH Key Creation (On Deploy Page)
SSH keys are created on the deployment page at barrack.ai/deploy:
- Go to barrack.ai/deploy
- Find the SSH key creation option on the deploy page
- Create a new SSH key (manual step - not automatic)
- Download starts - key saved to Downloads folder
- File format:
barrack_[keyname].pem
Step 2: Get Your Public IP
If Public IP enabled during deployment:
- Find your VM in dashboard
- Copy the Public IP (format:
185.x.x.x
)
If Public IP not enabled:
- Go to dashboard → Click VM “Details”
- Enable “Public IP” → Wait 1-2 minutes → Copy new IP
Step 3: Connect via SSH
Windows (Command Prompt/PowerShell):
Example:
Mac/Linux:
Always use Public IP (185.x.x.x), NOT Private IP (10.x.x.x). Username is always ubuntu
for all Barrack images.
Quick Deployment Guide
1. Choose Your GPU
Select from our high-performance GPU options:
GPU Model | Performance Level | Best For | Price (USD/hour) |
---|---|---|---|
H100-SXM5-80GB | Highest | Large language models, complex AI research | $2.69 |
H100-PCIe-NVLink-80GB | Very High | Multi-GPU workloads, distributed training | $2.29 |
H100-PCIe-80GB | Very High | Single-GPU inference, model training | $2.25 |
A100-SXM4-80GB-NVLink | High | Deep learning, scientific computing | $1.59 |
A100-PCIe-80GB | High | ML training, data analytics | $1.55 |
L40 | Medium-High | AI inference, graphics workloads | $1.19 |
RTX-A6000 | Medium | Development, small-scale training | $0.69 |
A40 | Medium | Professional graphics, AI development | $0.69 |
2. Select Region
Choose the geographic location closest to your users:
NORWAY-1
Europe Region
- Low latency for European users
- All GPU types available
CANADA-1
North America Region
- Optimized for North American users
- All GPU types available
US-1
United States Region
- US-based infrastructure
- All GPU types available
3. Configure Boot Source
Choose from multiple boot source options:
Pre-configured operating systems optimized for GPU workloads
Available images:
- Ubuntu Server 22.04 LTS R550 CUDA 12.4
- Ubuntu Server 22.04 LTS R550 CUDA 12.4 with Docker
- Ubuntu Server 20.04 LTS R535 CUDA 12.2
- Ubuntu Server 20.04 LTS R535 CUDA 12.2 with Docker
Pre-configured operating systems optimized for GPU workloads
Available images:
- Ubuntu Server 22.04 LTS R550 CUDA 12.4
- Ubuntu Server 22.04 LTS R550 CUDA 12.4 with Docker
- Ubuntu Server 20.04 LTS R535 CUDA 12.2
- Ubuntu Server 20.04 LTS R535 CUDA 12.2 with Docker
Boot from an existing storage volume (authenticated users only)
Requirements:
- Requires a bootable volume in the same region
- Volumes are filtered by region and bootable status
- Persistent between deployments
- Minimum 100GB for bootable volumes
Use your own saved VM images (authenticated users only)
Features:
- Created from your own snapshots
- Pre-configured environments
- Region-specific availability
- Standardize deployments across teams
Post-Deployment Steps
Once your VM is deployed:
Connect via SSH
Use your private key to securely connect to your VM
Configure Firewall Rules
Set up network security rules for your specific needs
Create Storage Volumes
Add persistent storage for your data and applications
Take Snapshots
Create backups for recovery and replication
VM deployment typically takes 2-5 minutes. You’ll receive notifications when your VM is ready.