FugokuFugoku Docs
Mask

Block Storage

Persistent volumes, snapshots, and backups for your instances

Block Storage

Fugoku Block Storage provides persistent, high-performance SSD volumes that can be attached to instances.

Overview

Key features:

  • Persistent: Data survives instance deletion
  • Expandable: Resize volumes without data loss
  • Triple-replicated: 99.999% durability
  • High IOPS: NVMe SSD backend
  • Snapshots: Point-in-time backups
  • Portable: Detach and reattach to different instances

Pricing

  • Storage: $0.10/GB/month
  • Snapshots: $0.05/GB/month
  • IOPS: Included (up to 20,000 IOPS per volume)

Examples:

100 GB volume:    $10/month
500 GB volume:    $50/month
1 TB volume:     $100/month
10 TB volume:  $1,000/month

Creating Volumes

Via Console

  1. Navigate to Storage → Volumes
  2. Click Create Volume
  3. Configure:
    • Name: my-data
    • Size: 100 GB (min 10 GB, max 10 TB)
    • Region: lagos-1 (must match instance)
    • Source: Blank or clone from snapshot
  4. Click Create

Volume ready in 10-30 seconds.

Via CLI

# Create blank volume
fugoku volumes create \
  --name my-data \
  --size 100 \
  --region lagos-1

# Create from snapshot
fugoku volumes create \
  --name restored-data \
  --size 100 \
  --region lagos-1 \
  --snapshot snap-abc123

Via API

curl -X POST https://api.fugoku.com/v1/volumes \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "my-data",
    "size_gb": 100,
    "region": "lagos-1"
  }'

Attaching Volumes

Volumes must be attached to an instance before use.

Via Console

  1. Volume detail page → Attach to Instance
  2. Select instance
  3. Volume appears as block device (e.g., /dev/vdb)

Via CLI

fugoku volumes attach my-data --instance web-1

Via API

curl -X POST https://api.fugoku.com/v1/volumes/my-data/attach \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"instance_id": "web-1"}'

Formatting and Mounting

After attaching, format and mount the volume.

First-Time Setup

# SSH into instance
fugoku ssh web-1

# List block devices
lsblk
# NAME   MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
# vda      8:0    0   80G  0 disk
# ├─vda1   8:1    0   79G  0 part /
# └─vda2   8:2    0    1G  0 part [SWAP]
# vdb      8:16   0  100G  0 disk

# Format the volume (ext4)
sudo mkfs.ext4 /dev/vdb

# Create mount point
sudo mkdir /mnt/data

# Mount volume
sudo mount /dev/vdb /mnt/data

# Verify
df -h /mnt/data
# Filesystem      Size  Used Avail Use% Mounted on
# /dev/vdb        98G   24K   93G   1% /mnt/data

Auto-Mount on Boot

Add to /etc/fstab:

# Get UUID (recommended over /dev/vdb)
sudo blkid /dev/vdb
# /dev/vdb: UUID="abc123-def456..." TYPE="ext4"

# Add to fstab
echo 'UUID=abc123-def456...  /mnt/data  ext4  defaults,nofail  0  2' | sudo tee -a /etc/fstab

# Test fstab
sudo mount -a

# Reboot and verify
sudo reboot
# (wait 30 seconds)
fugoku ssh web-1
df -h /mnt/data

Note: Use nofail option so boot succeeds even if volume isn't attached.

Alternative Filesystems

XFS (good for large files):

sudo apt install xfsprogs
sudo mkfs.xfs /dev/vdb
sudo mount /dev/vdb /mnt/data

Btrfs (advanced features):

sudo apt install btrfs-progs
sudo mkfs.btrfs /dev/vdb
sudo mount /dev/vdb /mnt/data

Detaching Volumes

Safely detach volumes for migration or maintenance.

Steps

# SSH into instance
fugoku ssh web-1

# Unmount volume
sudo umount /mnt/data

# Remove from fstab (optional, for permanent detach)
sudo nano /etc/fstab
# Comment out or delete the line for /dev/vdb

Via Console

Volume detail → Detach

Via CLI

fugoku volumes detach my-data

Note: Volume must be unmounted before detaching, or data corruption may occur.

Resizing Volumes

Expand volumes without data loss (shrinking not supported).

Resize Process

# Via CLI (can be done while attached)
fugoku volumes resize my-data --size 200

# Via Console
# Volume detail → Resize → Enter new size → Confirm

Expand Filesystem

After resizing, expand the filesystem:

# SSH into instance
fugoku ssh web-1

# For ext4
sudo resize2fs /dev/vdb

# For XFS
sudo xfs_growfs /mnt/data

# Verify
df -h /mnt/data
# Should show new size

Snapshots

Point-in-time backups of volumes.

Manual Snapshots

Via Console:

  1. Volume detail → Snapshots tab
  2. Click Create Snapshot
  3. Name snapshot
  4. Wait for completion (time depends on volume size and data changed)

Via CLI:

fugoku snapshots create my-data --name before-upgrade

Via API:

curl -X POST https://api.fugoku.com/v1/volumes/my-data/snapshots \
  -H "Authorization: Bearer $FUGOKU_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "before-upgrade"}'

Automatic Snapshots

Schedule recurring snapshots.

Via Console: Volume detail → Snapshots → Enable Automatic

  • Schedule: Hourly, Daily, Weekly
  • Retention: 1-365 days
  • Time: Preferred hours (for daily/weekly)

Via CLI:

fugoku snapshots enable \
  --volume my-data \
  --schedule daily \
  --time 02:00 \
  --retention 7

Example schedules:

  • Hourly, keep 24 hours
  • Daily at 2:00 AM, keep 7 days
  • Weekly on Sunday at 3:00 AM, keep 4 weeks

Restoring from Snapshots

Create new volume from snapshot:

fugoku volumes create \
  --name restored-data \
  --size 100 \
  --region lagos-1 \
  --snapshot snap-abc123

Overwrite existing volume:

  1. Detach volume from instance
  2. Volume detail → Restore tab
  3. Select snapshot
  4. Confirm (destroys current data)

Clone for testing:

Create copy of production data for staging:

# Snapshot production volume
fugoku snapshots create prod-db --name prod-snapshot

# Create staging volume from snapshot
fugoku volumes create \
  --name staging-db \
  --size 100 \
  --region lagos-1 \
  --snapshot prod-snapshot

# Attach to staging instance
fugoku volumes attach staging-db --instance staging-db-1

Performance

IOPS & Throughput

All volumes use NVMe SSDs with high performance:

  • Random Read: Up to 20,000 IOPS
  • Random Write: Up to 10,000 IOPS
  • Sequential Read: Up to 500 MB/s
  • Sequential Write: Up to 250 MB/s

Performance scales with volume size up to limits above.

Benchmarking

Test IOPS (random reads):

sudo apt install fio

# Random read test
sudo fio --name=randread \
  --ioengine=libaio \
  --iodepth=32 \
  --rw=randread \
  --bs=4k \
  --direct=1 \
  --size=1G \
  --numjobs=4 \
  --runtime=60 \
  --group_reporting \
  --filename=/mnt/data/testfile

Test throughput (sequential writes):

sudo fio --name=seqwrite \
  --ioengine=libaio \
  --iodepth=32 \
  --rw=write \
  --bs=128k \
  --direct=1 \
  --size=4G \
  --numjobs=1 \
  --runtime=60 \
  --group_reporting \
  --filename=/mnt/data/testfile

Expected results:

Random read:   15,000-20,000 IOPS
Sequential write:  200-250 MB/s

Optimization Tips

Use noatime mount option: Reduces write load by not updating access times.

# In /etc/fstab
UUID=abc123...  /mnt/data  ext4  defaults,noatime,nofail  0  2

Tune I/O scheduler: For database workloads, use noop or deadline scheduler.

echo deadline | sudo tee /sys/block/vdb/queue/scheduler

Increase readahead: For sequential workloads.

sudo blockdev --setra 2048 /dev/vdb

Use Cases

Database Storage

Separate data from OS for easier management:

# Create volume for PostgreSQL
fugoku volumes create \
  --name postgres-data \
  --size 200 \
  --region lagos-1

# Attach to database instance
fugoku volumes attach postgres-data --instance db-1

# Format and mount
sudo mkfs.ext4 /dev/vdb
sudo mkdir /var/lib/postgresql
sudo mount /dev/vdb /var/lib/postgresql
echo '/dev/vdb /var/lib/postgresql ext4 defaults,noatime,nofail 0 2' | sudo tee -a /etc/fstab

# Install PostgreSQL
sudo apt install postgresql

# Data automatically stored on volume

File Storage

Shared files across multiple applications:

# Volume for uploads/media
fugoku volumes create \
  --name app-uploads \
  --size 500 \
  --region lagos-1

# Attach to web server
fugoku volumes attach app-uploads --instance web-1

# Mount
sudo mkfs.ext4 /dev/vdb
sudo mkdir /var/www/uploads
sudo mount /dev/vdb /var/www/uploads
sudo chown www-data:www-data /var/www/uploads

Backup Storage

Dedicated volume for backups:

# Large volume for backups
fugoku volumes create \
  --name backups \
  --size 1000 \
  --region lagos-1

# Attach to backup server
fugoku volumes attach backups --instance backup-1

# Mount
sudo mkfs.ext4 /dev/vdb
sudo mkdir /mnt/backups
sudo mount /dev/vdb /mnt/backups

# Run backup script
#!/bin/bash
BACKUP_DIR="/mnt/backups/$(date +%Y-%m-%d)"
mkdir -p "$BACKUP_DIR"
pg_dump production_db > "$BACKUP_DIR/database.sql"
tar czf "$BACKUP_DIR/files.tar.gz" /var/www

Development Environments

Preserve dev work across instance rebuilds:

# Volume for project code
fugoku volumes create \
  --name dev-projects \
  --size 100 \
  --region lagos-1

# Attach to dev instance
fugoku volumes attach dev-projects --instance dev-1

# Mount at home directory
sudo mkfs.ext4 /dev/vdb
sudo mkdir /home/ubuntu/projects
sudo mount /dev/vdb /home/ubuntu/projects
sudo chown ubuntu:ubuntu /home/ubuntu/projects

Data Migration

Between Instances

Move volume from one instance to another:

# Detach from source
fugoku ssh web-1
sudo umount /mnt/data
exit

fugoku volumes detach my-data

# Attach to destination
fugoku volumes attach my-data --instance web-2

# Mount on destination
fugoku ssh web-2
sudo mkdir /mnt/data
sudo mount /dev/vdb /mnt/data

Between Regions

Use snapshots to migrate data:

# In source region (lagos-1)
fugoku snapshots create my-data --name for-migration

# Copy snapshot to target region (coming Q3 2026)
fugoku snapshots copy snap-abc123 --to-region london-1

# In target region, create volume from snapshot
fugoku volumes create \
  --name my-data-london \
  --size 100 \
  --region london-1 \
  --snapshot snap-abc123

Current workaround (until snapshot copy available):

# Rsync over network
fugoku ssh web-lagos
rsync -avz /mnt/data/ ubuntu@web-london:/mnt/data/

Monitoring

Via Console

Volume detail → Metrics tab:

  • Read IOPS
  • Write IOPS
  • Read throughput (MB/s)
  • Write throughput (MB/s)
  • Latency

Via CLI

fugoku volumes stats my-data

# Output:
# Read:  1,200 IOPS,  45 MB/s
# Write:   800 IOPS,  30 MB/s
# Latency: 0.5 ms avg

Disk Space Monitoring

On instance:

# Check space
df -h /mnt/data

# Set up alert when 80% full
sudo apt install nagios-plugins-basic
check_disk -w 20% -c 10% -p /mnt/data

Security

Encryption

All volumes are encrypted at rest (AES-256).

  • Default: Fugoku-managed keys
  • Custom keys (coming Q4 2026): Bring your own encryption keys

Access Control

Volumes can only be attached to instances in your account.

Team permissions:

  • Admin: Full control
  • Developer: Attach/detach, create snapshots
  • Read-Only: View only

Backups

Enable automatic snapshots for critical volumes:

fugoku snapshots enable \
  --volume my-data \
  --schedule daily \
  --retention 30

Follow 3-2-1 rule:

  • 3 copies of data
  • 2 different storage types
  • 1 off-site backup

Troubleshooting

Volume won't attach

  • Check region: Volume and instance must be in same region
  • Check status: Volume must be "available" (not already attached)
  • Retry: Detach and reattach

Can't mount volume

# Check if volume is attached
lsblk

# Check filesystem
sudo file -s /dev/vdb
# Should show filesystem type

# Force check (if ext4)
sudo e2fsck -f /dev/vdb

Performance issues

  • Check IOPS limit: May be hitting 20k IOPS cap
  • I/O wait: Run top and check %wa - high I/O wait indicates bottleneck
  • Test with fio: Benchmark to isolate issue
top
# %wa > 10% indicates I/O bottleneck

# Check I/O stats
iostat -x 1

Snapshot failed

  • Volume may be very active - pause writes temporarily
  • Check disk space on volume
  • Contact support if persists: support@fugoku.com

Best Practices

  1. Use volumes for persistent data - root disks are ephemeral if instance deleted
  2. Enable automatic snapshots - daily backups for critical data
  3. Test restores - verify snapshots work before disaster strikes
  4. Monitor disk space - set up alerts at 80% full
  5. Use appropriate filesystem - ext4 for general use, XFS for large files
  6. Label volumes clearly - use descriptive names (prod-db, staging-uploads)
  7. Document mount points - keep inventory of what's mounted where

Limits

Per account:

  • Volumes: 100 per region
  • Total storage: 50 TB per region
  • Snapshots: 500 per account

Need higher limits? Contact support@fugoku.com

Pricing Calculator

Volume:  100 GB × $0.10 = $10.00/month
Snapshot: 50 GB × $0.05 =  $2.50/month
────────────────────────────────────
Total:                     $12.50/month

Snapshots use incremental storage (only changed blocks), so 10 snapshots of 100 GB volume != 1 TB cost.

Next Steps:

GPU Compute

NVIDIA H100 and A100 GPUs for AI/ML training, inference, and high-performance computing

Networking

Private networks, floating IPs, load balancers, and firewalls

On this page

Block Storage
Overview
Pricing
Creating Volumes
Via Console
Via CLI
Via API
Attaching Volumes
Via Console
Via CLI
Via API
Formatting and Mounting
First-Time Setup
Auto-Mount on Boot
Alternative Filesystems
Detaching Volumes
Steps
Via Console
Via CLI
Resizing Volumes
Resize Process
Expand Filesystem
Snapshots
Manual Snapshots
Automatic Snapshots
Restoring from Snapshots
Performance
IOPS & Throughput
Benchmarking
Optimization Tips
Use Cases
Database Storage
File Storage
Backup Storage
Development Environments
Data Migration
Between Instances
Between Regions
Monitoring
Via Console
Via CLI
Disk Space Monitoring
Security
Encryption
Access Control
Backups
Troubleshooting
Volume won't attach
Can't mount volume
Performance issues
Snapshot failed
Best Practices
Limits
Pricing Calculator