Compute Instances

Fugoku compute instances are virtual machines running on dedicated CPU cores with guaranteed RAM and SSD storage.

Plans & Pricing

All plans include:

Dedicated vCPU (no CPU stealing)
Guaranteed RAM (no overcommitment)
NVMe SSD storage
1 Gbps network connection
1 TB bandwidth/month included
Unlimited internal traffic between instances

Available Plans

Plan	Cores / vCPU	RAM	Storage	Network	Hourly	Monthly	Yearly
`vm-standard`	4	16 GB	80 GB NVMe	1 Gbps	$0.35	$254	$2,124
`m4.metal.small`	6	64 GB	2×960 GB NVMe	10 Gbps	$0.41	$297	$2,487
`f4.metal.small`	12	96 GB	2×960 GB NVMe	10 Gbps	$0.55	$399	$3,345
`m4.metal.large`	24	384 GB	4×NVMe	—	$2.03	$1,471	$12,324
`rs4.metal.large`	32	768 GB	2×480 GB + 2×8 TB NVMe	100 Gbps	$3.22	$2,351	$19,748
`m4.metal.xlarge`	48	768 GB	4×NVMe	—	$4.12	$2,986	$25,024
`rs4.metal.xlarge`	64	1536 GB	4×8 TB NVMe	100 Gbps	$6.44	$4,669	$39,110
`f4.metal.large-gpu`	24	768 GB	2×480 GB + 2×3.8 TB NVMe	8× A100	$8.99	$6,521	$54,600

See the Servers page for detailed specs and features per plan.

Operating System Images

Linux Distributions

All images are official upstream builds, updated monthly.

Ubuntu:

ubuntu-22.04 (Jammy Jellyfish) - LTS, supported until 2027
ubuntu-24.04 (Noble Numbat) - LTS, supported until 2029
ubuntu-24.10 (Oracular Oriole) - Latest stable

Debian:

debian-11 (Bullseye) - Stable
debian-12 (Bookworm) - Stable
debian-testing (Trixie) - For adventurous users

CentOS / Rocky Linux:

rocky-8 - RHEL 8 compatible
rocky-9 - RHEL 9 compatible
almalinux-8 - Alternative RHEL 8 rebuild
almalinux-9 - Alternative RHEL 9 rebuild

Fedora:

fedora-39 - Cutting edge
fedora-40 - Latest stable

Alpine:

alpine-3.18 - Minimal (5 MB base image)
alpine-3.19 - Latest

Arch Linux:

arch - Rolling release (for the brave)

Default Credentials

Image	Default User	Auth Method
Ubuntu/Debian	ubuntu / debian	SSH key only
Rocky/Alma	rocky / almalinux	SSH key only
Fedora	fedora	SSH key only
Alpine	root	SSH key only
Arch	arch	SSH key only

Password auth is disabled by default for security. Use SSH keys.

Creating an Instance

Via Console

Navigate to Servers
Click Create Server
Select Virtual Machine category
Select provider, region, and plan
Configure name, SSH keys, user data
Click Create

Provisioning time: 30-90 seconds.

Via CLI

# Basic creation
fugoku create server \
  --name web-1 \
  --plan vm-standard \
  --image ubuntu-22.04 \
  --region lagos-1 \
  --role VirtualMachine

# With custom configuration
fugoku create server \
  --name api-prod \
  --plan m4.metal.small \
  --image ubuntu-22.04 \
  --region lagos-1 \
  --role VirtualMachine \
  --ssh-key my-laptop \
  --private-network backend-net \
  --enable-backups \
  --tags env:production,team:backend \
  --user-data @cloud-init.yaml

Via API (Legacy)

curl -X POST https://api.fugoku.com/v1/instances \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "web-1",
    "region": "lagos-1",
    "instanceType": "vm-standard",
    "image": "ubuntu-22.04",
    "ssh_keys": ["my-laptop"],
    "tags": ["env:production"],
    "backups": true
  }'

Via API (Recommended)

curl -X POST https://api.fugoku.com/v1/servers \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "data": {
      "type": "servers",
      "attributes": {
        "name": "web-1",
        "hostname": "web-1.example.com",
        "site": "lagos-1",
    "instanceType": "vm-standard",
        "operating_system": "ubuntu-22.04",
        "billing": "hourly",
        "role": "VirtualMachine",
        "provider": "latitude",
        "ssh_keys": ["my-laptop"],
        "tags": [{ "key": "env", "value": "production" }]
      }
    }
  }'

Cloud-Init / User Data

Automate instance setup with cloud-init scripts.

Example: Install Nginx and deploy app

#cloud-config
package_update: true
package_upgrade: true

packages:
  - nginx
  - git
  - nodejs
  - npm

runcmd:
  - systemctl start nginx
  - systemctl enable nginx
  - cd /var/www
  - git clone https://github.com/yourname/app.git
  - cd app && npm install
  - npm start &

write_files:
  - path: /etc/nginx/sites-available/app
    content: |
      server {
        listen 80;
        server_name _;
        location / {
          proxy_pass http://localhost:3000;
          proxy_http_version 1.1;
          proxy_set_header Upgrade $http_upgrade;
          proxy_set_header Connection 'upgrade';
          proxy_set_header Host $host;
          proxy_cache_bypass $http_upgrade;
        }
      }
  - path: /etc/systemd/system/app.service
    content: |
      [Unit]
      Description=Node.js App
      After=network.target

      [Service]
      Type=simple
      User=www-data
      WorkingDirectory=/var/www/app
      ExecStart=/usr/bin/node server.js
      Restart=on-failure

      [Install]
      WantedBy=multi-user.target

runcmd:
  - ln -s /etc/nginx/sites-available/app /etc/nginx/sites-enabled/
  - rm /etc/nginx/sites-enabled/default
  - nginx -t && systemctl reload nginx
  - systemctl daemon-reload
  - systemctl start app
  - systemctl enable app

Pass this to --user-data:

fugoku create server \
  --name web-1 \
  --plan vm-standard \
  --image ubuntu-22.04 \
  --region lagos-1 \
  --role VirtualMachine \
  --user-data @setup.yaml

SSH Access

Upload SSH Keys

Console: Account → SSH Keys → Add Key

CLI:

fugoku ssh-keys add --name laptop ~/.ssh/id_rsa.pub

API:

curl -X POST https://api.fugoku.com/v1/account/ssh-keys \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "laptop",
    "public_key": "ssh-rsa AAAAB3NzaC1yc2E... user@laptop"
  }'

Connect to Instance

Get IP address:

fugoku instances get web-1
# IP: 102.89.45.178

SSH directly:

# With CLI (manages keys automatically)
fugoku ssh web-1

# Manual SSH
ssh ubuntu@102.89.45.178

Troubleshooting SSH

Connection refused:

Instance may still be provisioning (wait 60 seconds)
Check firewall rules (port 22 must be open)

# View firewall rules
fugoku firewalls list-rules web-1

# Add SSH rule if missing
fugoku firewalls add-rule web-1 \
  --protocol tcp \
  --port 22 \
  --source your.ip.address/32

Permission denied:

Verify SSH key is uploaded to your account
Use correct username for image (ubuntu, debian, etc.)
Check key permissions: chmod 600 ~/.ssh/id_rsa

Managing Instances

Start / Stop / Reboot

Console: Server detail → Actions menu

CLI:

# Stop instance (free while stopped, data preserved)
fugoku servers stop web-1

# Start instance
fugoku servers start web-1

# Reboot instance
fugoku servers reboot web-1

API:

curl -X POST https://api.fugoku.com/v1/servers/web-1/actions \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "data": {
      "type": "actions",
      "attributes": { "action": "power_off" }
    }
  }'

Billing note: Stopped instances only incur storage charges (~$0.10/GB/mo).

Resize Instance

Change CPU/RAM (requires reboot):

Console: Server detail → Upgrade tab → Select new plan → Confirm

CLI:

fugoku servers upgrade web-1 --plan m4.metal.small

Process:

Server status changes to migrating
Plan changes
Server restarts with new resources
Downtime: ~60 seconds

Rebuild Instance

Reinstall OS (destroys all data, keeps IP):

Console: Server detail → Actions → Rebuild

CLI:

fugoku servers rebuild web-1 --image ubuntu-24.04

Use cases:

Start fresh after compromise
Switch to different OS
Reset to clean state

Delete Instance

Permanently destroy instance and all data.

Console: Server detail → Actions → Delete

CLI:

fugoku instances delete web-1 --confirm

Recovery: Deleted instances are not recoverable. Take snapshot first.

Monitoring & Metrics

View Metrics

Console: Server detail → Metrics tab

CLI:

fugoku instances stats web-1

# Output:
# CPU: 12.5%
# RAM: 1.8 GB / 4 GB (45%)
# Disk: 2.1 GB / 80 GB (3%)
# Network: ↓ 450 Kbps ↑ 120 Kbps

API:

curl https://api.fugoku.com/v1/servers/web-1/metrics?period=1h \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN"

Set Up Alerts

Console: Server detail → Alerts → Create Alert Rule

Example rules:

CPU >80% for 5 minutes → email
RAM >90% for 10 minutes → email + SMS
Disk >95% full → email
Network unreachable for 2 minutes → email

CLI:

fugoku alerts create \
  --instance web-1 \
  --metric cpu \
  --threshold 80 \
  --duration 5m \
  --notify email:support@yourcompany.com

Backups & Snapshots

Manual Backups

Console: Server detail → Backups tab → Create backup

CLI:

fugoku backups create web-1 --name before-upgrade

Restore:

fugoku backups restore backup-abc123

Automatic Backups

Console: Server detail → Backups tab → Enable Automatic

CLI:

fugoku backups enable \
  --instance web-1 \
  --schedule daily \
  --time 02:00 \
  --retention 7

Schedule options:

hourly (keep last N hours)
daily (keep last N days)
weekly (keep last N weeks)

Cost: $0.05/GB/month for backup storage

Networking

Public IP

Every instance gets a public IPv4 by default.

View IP:

fugoku instances get web-1 | grep ip

Floating IP (optional): Static IP that persists across rebuilds and can move between instances.

# Create floating IP
fugoku networking create-ip --region lagos-1

# Assign to instance
fugoku networking assign-ip web-1 --ip 102.89.45.200

Private Networking

Connect instances on isolated VLANs.

Create network:

fugoku networks create \
  --name backend \
  --subnet 10.10.0.0/24 \
  --region lagos-1

Attach instance:

fugoku networks attach backend --instance web-1

Instance gets second interface: eth1 with private IP 10.10.0.x

Use case: Database servers only accessible from app servers, not internet.

Performance Tips

CPU Performance

Use CPU-optimized plans for compute-heavy tasks
Monitor CPU steal (should be <1%) - run top and check %st
Fugoku uses dedicated cores, not shared

Disk I/O

Root disk is NVMe SSD (fast)
For very high IOPS, use storage-optimized plans
Attach block volumes for additional storage

Benchmark disk:

# Write test
dd if=/dev/zero of=/tmp/test bs=1M count=1024 oflag=direct
# Should see 500+ MB/s on NVMe

# Read test
dd if=/tmp/test of=/dev/null bs=1M iflag=direct

Network Performance

Internal traffic between Fugoku instances: 10 Gbps, free
Public internet: 1 Gbps, 1 TB/month included
Additional bandwidth: $0.01/GB

Benchmark network:

# Install iperf3
sudo apt install iperf3

# On server instance
iperf3 -s

# On client instance
iperf3 -c server-ip
# Should see 900+ Mbps

Security Best Practices

Disable password auth, use SSH keys only
Use firewall rules - restrict port 22 to your IP

Enable automatic updates:

sudo apt install unattended-upgrades
sudo dpkg-reconfigure -plow unattended-upgrades

Use private networks for internal services
Enable backups before major changes
Rotate SSH keys quarterly

Run fail2ban to block brute force attempts:

sudo apt install fail2ban
sudo systemctl start fail2ban
sudo systemctl enable fail2ban

Cost Optimization

Stop unused instances: Stopped instances only cost ~$0.10/GB/month for storage.

Right-size plans: Monitor metrics and downgrade if CPU/RAM underutilized.

Use reservations (coming soon): Commit to 1-3 year term for 30-50% discount.

Spot instances (coming Q3 2026): Spare capacity at up to 80% discount for fault-tolerant workloads.

Troubleshooting

Instance won't start:

Check account balance
Verify region has capacity
View error logs in Console

Poor performance:

Check metrics for bottlenecks
Verify plan size is adequate
Review disk I/O and network usage
Contact support if CPU steal >1%

Can't connect:

Verify firewall rules
Check instance is running
Confirm SSH key is uploaded
Try web console (VNC)

Getting Help

Documentation: docs.fugoku.com
Support: support@fugoku.com
Status: status.fugoku.com
Discord: discord.gg/fugoku

Next Steps:

Explore GPU Compute for AI/ML workloads
Set Up Block Storage for additional capacity
Configure Private Networks for isolation
Use the API for automation

Compute Instances

On this page