Skip to main content

Command Palette

Search for a command to run...

Day 4 : Create an GP3 Volume

Updated
5 min read
Day 4 : Create an GP3 Volume

AWS EBS Volumes – Concept, Usage, and Industry Practices

1. Problem Statement

A task was assigned to create an AWS volume with a specific name, type, and size.
While the task looks simple, misunderstanding AWS volume fundamentals can lead to:

  • Incorrect volume type selection

  • Performance bottlenecks

  • Increased AWS costs

  • Failed EC2 attachments due to AZ mismatch

  • Poor production and audit outcomes

This document explains what an AWS volume is, how it works, why it exists, and how it is used in real enterprise environments.


2. Can an AWS Volume Be Created Independently?

Yes.
An Amazon EBS volume can be created independently without attaching it to an EC2 instance.

Key points:

  • EBS volumes are created at the Availability Zone level

  • A volume can exist unattached

  • Charges apply even if the volume is unused

  • A volume can only be attached to EC2 instances in the same AZ

Common mistake:
Creating a volume in one AZ and attempting to attach it to an EC2 in another AZ will fail.


3. Should Volumes Be Created Without a Use Case?

Technically yes.
Operationally no.

In real environments, volumes are created for:

  • Application data storage

  • Database persistence

  • Log storage

  • Migration activities

  • Disaster recovery via snapshots

  • Automation pipelines

Creating volumes without purpose leads to:

  • Wasted cost

  • Audit findings

  • Operational confusion


4. Types of AWS EBS Volumes You Must Know

These five are mandatory knowledge.

4.1 gp3 (General Purpose SSD)

  • Default and recommended

  • Balanced performance and cost

  • IOPS and throughput are configurable independently

  • Used for:

    • Application servers

    • Root volumes

    • Most workloads

4.2 gp2 (Legacy General Purpose SSD)

  • IOPS tied to volume size

  • Older generation

  • Still present in legacy systems

4.3 io1 / io2 (Provisioned IOPS SSD)

  • High and consistent IOPS

  • Very expensive

  • Used for:

    • Databases

    • Mission critical workloads

    • Low latency systems

4.4 st1 (Throughput Optimized HDD)

  • High throughput for sequential access

  • Cost effective

  • Used for:

    • Big data

    • Log processing

    • Streaming workloads

4.5 sc1 (Cold HDD)

  • Lowest cost

  • Lowest performance

  • Used for:

    • Archival data

    • Rarely accessed workloads


5. Factors to Consider When Selecting a Volume

Always evaluate in this order.

5.1 Workload Pattern

  • Random IO or sequential IO

  • Read intensive or write intensive

5.2 Performance Requirements

  • Required IOPS

  • Required throughput

  • Latency sensitivity

5.3 Cost Considerations

  • SSD vs HDD

  • Provisioned vs burstable performance

5.4 Durability and Recovery

  • Snapshot requirements

  • Backup frequency

  • Restore time objectives

5.5 Scalability

  • Ability to resize volume

  • Ability to modify IOPS without downtime

Wrong selection causes outages or cost overruns.


6. Common Interview Questions on AWS Volumes

You must be able to answer these confidently.

  • Difference between gp2 and gp3

  • Can an EBS volume be attached to multiple EC2 instances

  • What happens to EBS when EC2 is terminated

  • Difference between EBS and Instance Store

  • What is an EBS snapshot and where it is stored

  • How to migrate data between volumes

  • Can volume type be changed without downtime

Hesitation here indicates weak fundamentals.


7. Real Industry Tasks Involving Volumes

7.1 Daily Operations

  • Creating and attaching volumes

  • Expanding volume size

  • Modifying IOPS and throughput

7.2 Production Support

  • Restoring from snapshots

  • Replacing corrupted volumes

  • Migrating volumes during incidents

7.3 Cost Optimization

  • Migrating gp2 to gp3

  • Identifying and deleting unattached volumes

  • Right sizing storage

7.4 Security and Compliance

  • Enabling encryption

  • Using KMS-managed keys

  • Auditing volume access

7.5 Automation

  • Volume provisioning using Terraform or CloudFormation

  • Auto attachment in Auto Scaling

  • Snapshot lifecycle management


8. Key Takeaway

EBS volumes are not just storage units.
They directly impact performance, reliability, cost, and recovery.

Question and answers:

1. Difference between gp2 and gp3

gp2

  • Performance depends on volume size

  • Bigger volume means more IOPS

  • Older generation

  • Less cost-efficient

gp3

  • Performance and size are independent

  • You can set IOPS and throughput separately

  • Newer and cheaper

  • Better control and predictable performance

Bottom line:
gp3 is better. gp2 exists mainly for legacy systems.


2. Can an EBS volume be attached to multiple EC2 instances?

Normally: No

  • One EBS volume attaches to one EC2 instance

  • Exception: io1 and io2 support multi-attach

  • Multi-attach is rare and used only for special clustered apps

Industry reality:
99 percent of systems use one volume per EC2.


3. What happens to EBS when EC2 is terminated?

It depends on the volume type.

  • Root volume

    • Deleted by default when EC2 is terminated
  • Additional volumes

    • Not deleted by default

    • Remain in AWS and continue to cost money

This causes hidden bills if ignored.


4. Difference between EBS and Instance Store

EBS

  • Persistent storage

  • Data survives EC2 stop or start

  • Can be detached and attached to another EC2

  • Slower than instance store but reliable

Instance Store

  • Temporary storage

  • Data lost when EC2 stops or terminates

  • Very fast

  • Used for cache or temporary processing

Rule:
Important data never goes to instance store.


5. What is an EBS snapshot and where is it stored?

  • A snapshot is a backup of an EBS volume

  • Stored in Amazon S3

  • Incremental by default

  • Used for:

    • Backup

    • Disaster recovery

    • Creating new volumes

Snapshots are region-specific.


6. How to migrate data between volumes?

Simple and common methods:

  • Attach both volumes to the same EC2

  • Use rsync, cp, or dd

  • Create a snapshot and restore it to a new volume

  • Detach old volume and attach new volume

This is routine work in production.


7. Can volume type be changed without downtime?

Yes, in most cases

  • AWS supports live volume modification

  • You can change:

    • Volume type

    • Size

    • IOPS

    • Throughput

But

  • Filesystem resize may be needed

  • Performance may degrade briefly

  • Always test in non-prod first