EC2 EBS Volumes

Amazon Elastic Block Store (EBS) provides persistent block storage volumes for EC2 instances. EBS volumes persist independently from the running life of an instance.

Key Benefits

EBS volumes are automatically replicated within their Availability Zone to protect against hardware failure, providing high availability and durability.

Volume Types

gp3 (General Purpose SSD)

The latest generation general-purpose SSD. Recommended for most workloads.

Attribute	Value
Size	1 GiB - 16 TiB
Baseline IOPS	3,000
Max IOPS	16,000
Baseline Throughput	125 MiB/s
Max Throughput	1,000 MiB/s

Create gp3 Volume

aws ec2 create-volume \
  --volume-type gp3 \
  --size 100 \
  --iops 4000 \
  --throughput 250 \
  --availability-zone us-east-1a

io2 Block Express (Provisioned IOPS)

Highest performance SSD for critical workloads.

Attribute	Value
Size	4 GiB - 64 TiB
Max IOPS	256,000
Max Throughput	4,000 MiB/s
Durability	99.999%

Create io2 Volume

aws ec2 create-volume \
  --volume-type io2 \
  --size 500 \
  --iops 64000 \
  --availability-zone us-east-1a

st1 (Throughput Optimized HDD)

Low-cost HDD for frequently accessed, throughput-intensive workloads.

Attribute	Value
Size	125 GiB - 16 TiB
Max IOPS	500
Max Throughput	500 MiB/s

Best for: Big data, data warehouses, log processing

sc1 (Cold HDD)

Lowest cost HDD for less frequently accessed workloads.

Attribute	Value
Size	125 GiB - 16 TiB
Max IOPS	250
Max Throughput	250 MiB/s

Best for: Infrequently accessed data, archives

Type	Use Case	IOPS	Throughput	Cost
gp3	Most workloads	16,000	1,000 MiB/s	$$
gp2	Legacy general purpose	16,000	250 MiB/s	$$
io2	Critical databases	256,000	4,000 MiB/s	$$$$
io1	Legacy high IOPS	64,000	1,000 MiB/s	$$$
st1	Big data	500	500 MiB/s	$
sc1	Cold storage	250	250 MiB/s	$

For most workloads, gp3 offers the best price/performance. Use io2 only for demanding database workloads.

Managing Volumes

Create and Attach

Create Volume

aws ec2 create-volume \
  --volume-type gp3 \
  --size 100 \
  --availability-zone us-east-1a \
  --tag-specifications 'ResourceType=volume,Tags=[{Key=Name,Value=data-volume}]'

Attach to Instance

Attach Volume

aws ec2 attach-volume \
  --volume-id vol-0123456789abcdef0 \
  --instance-id i-0123456789abcdef0 \
  --device /dev/xvdf

Format and Mount (on instance)

Format and Mount

# Check device name (might be different from specified)
lsblk

# Create filesystem
sudo mkfs -t xfs /dev/xvdf

# Create mount point
sudo mkdir /data

# Mount
sudo mount /dev/xvdf /data

# Add to fstab for persistence
echo '/dev/xvdf /data xfs defaults,nofail 0 2' | sudo tee -a /etc/fstab

Modify Volume

Increase size or change type without downtime:

Modify Volume

aws ec2 modify-volume \
  --volume-id vol-0123456789abcdef0 \
  --size 200 \
  --volume-type gp3 \
  --iops 6000 \
  --throughput 400

After modification, extend the filesystem:

Extend Filesystem

# For XFS
sudo xfs_growfs /data

# For ext4
sudo resize2fs /dev/xvdf

You can only increase volume size, not decrease. There's a 6-hour cooldown between modifications.

Detach and Delete

Detach Volume

# Unmount on instance first
sudo umount /data

# Detach
aws ec2 detach-volume --volume-id vol-0123456789abcdef0

# Delete (careful - data is lost!)
aws ec2 delete-volume --volume-id vol-0123456789abcdef0

Snapshots

Snapshots are point-in-time copies of EBS volumes, stored in S3.

Create Snapshot

aws ec2 create-snapshot \
  --volume-id vol-0123456789abcdef0 \
  --description "Daily backup" \
  --tag-specifications 'ResourceType=snapshot,Tags=[{Key=Name,Value=daily-backup}]'

Restore from Snapshot

Create Volume from Snapshot

aws ec2 create-volume \
  --snapshot-id snap-0123456789abcdef0 \
  --volume-type gp3 \
  --availability-zone us-east-1a

Automate with Data Lifecycle Manager

Create Lifecycle Policy

aws dlm create-lifecycle-policy \
  --description "Daily EBS snapshots" \
  --state ENABLED \
  --execution-role-arn arn:aws:iam::123456789012:role/AWSDataLifecycleManagerDefaultRole \
  --policy-details '{
    "PolicyType": "EBS_SNAPSHOT_MANAGEMENT",
    "ResourceTypes": ["VOLUME"],
    "TargetTags": [{"Key": "Backup", "Value": "true"}],
    "Schedules": [{
      "Name": "Daily",
      "CreateRule": {"Interval": 24, "IntervalUnit": "HOURS", "Times": ["03:00"]},
      "RetainRule": {"Count": 7},
      "CopyTags": true
    }]
  }'

Encryption

Enable Encryption

Create Encrypted Volume

aws ec2 create-volume \
  --volume-type gp3 \
  --size 100 \
  --encrypted \
  --kms-key-id alias/aws/ebs \
  --availability-zone us-east-1a

You cannot directly encrypt an existing unencrypted volume. Create an encrypted copy:

Create Snapshot

aws ec2 create-snapshot --volume-id vol-unencrypted123

Copy with Encryption

aws ec2 copy-snapshot \
  --source-region us-east-1 \
  --source-snapshot-id snap-unencrypted123 \
  --encrypted \
  --kms-key-id alias/aws/ebs

Create New Volume

aws ec2 create-volume \
  --snapshot-id snap-encrypted456 \
  --availability-zone us-east-1a

Enable default encryption for all new volumes in a region:

Enable Default Encryption

aws ec2 enable-ebs-encryption-by-default

# Check status
aws ec2 get-ebs-encryption-by-default

EBS encryption uses AWS KMS. There's no performance impact with encrypted volumes on modern instance types.

Multi-Attach

Allow io1/io2 volumes to attach to multiple instances (up to 16):

Create Multi-Attach Volume

aws ec2 create-volume \
  --volume-type io2 \
  --size 100 \
  --iops 10000 \
  --multi-attach-enabled \
  --availability-zone us-east-1a

Requirements:

io1 or io2 volumes only
Nitro-based instances only
Instances must be in same AZ
Requires cluster-aware filesystem (not ext4/xfs)

Performance Optimization

IOPS and Throughput

gp3 allows independent provisioning:

Provision IOPS and Throughput

aws ec2 modify-volume \
  --volume-id vol-0123456789abcdef0 \
  --iops 10000 \
  --throughput 500

EBS-Optimized Instances

Most modern instances are EBS-optimized by default:

Check EBS Optimization

aws ec2 describe-instance-types \
  --instance-types m6i.large \
  --query 'InstanceTypes[0].EbsInfo'

RAID for Higher Performance

For performance beyond single volume limits:

Configuration	Use Case
RAID 0	Maximum performance (no redundancy)
RAID 1	Mirroring (not recommended, use EBS replication)
RAID 10	Performance + redundancy

RAID 5/6 is not recommended due to parity write penalty.

Monitoring

CloudWatch Metrics

Metric	Description
`VolumeReadOps`	Read operations
`VolumeWriteOps`	Write operations
`VolumeReadBytes`	Bytes read
`VolumeWriteBytes`	Bytes written
`VolumeTotalReadTime`	Read latency
`VolumeTotalWriteTime`	Write latency
`VolumeQueueLength`	I/O queue length
`BurstBalance`	gp2 burst credit balance

Get Volume Metrics

aws cloudwatch get-metric-statistics \
  --namespace AWS/EBS \
  --metric-name VolumeReadOps \
  --dimensions Name=VolumeId,Value=vol-0123456789abcdef0 \
  --start-time 2024-01-01T00:00:00Z \
  --end-time 2024-01-02T00:00:00Z \
  --period 3600 \
  --statistics Sum

Best Practices

EBS Best Practices

Use gp3 for most workloads (better price/performance than gp2)
Enable encryption by default for all new volumes
Take regular snapshots using Data Lifecycle Manager
Right-size volumes - you can always increase later
Monitor performance - watch for IOPS/throughput limits
Use separate volumes for OS and data for easier management
Delete unused volumes - you pay for provisioned storage

Volume Types

gp3 (General Purpose SSD)

io2 Block Express (Provisioned IOPS)

st1 (Throughput Optimized HDD)

sc1 (Cold HDD)

Managing Volumes

Create and Attach

Create Volume

Attach to Instance

Format and Mount (on instance)

Modify Volume

Detach and Delete

Snapshots

Create Snapshot

Restore from Snapshot

Automate with Data Lifecycle Manager

Encryption

Enable Encryption

Create Snapshot

Copy with Encryption

Create New Volume

Multi-Attach

Performance Optimization

IOPS and Throughput

EBS-Optimized Instances

RAID for Higher Performance

Monitoring

CloudWatch Metrics

Best Practices

Next Steps

EC2 AMIs

EC2 CLI

On this page