Everything you need to know about Amazon EBS
If you’re looking for a user-friendly and cloud-based block storage system, Amazon EBS is the right solution for you. The system integrates with the entire AWS platform, and you’ll be able to use built-in Amazon security, backup and access control features. EBS provides you with four types of volumes for SSD-backed volumes and HDD-backed volumes.
Read on to learn what is EBS, how it works, and what use cases are most suitable for this storage system.
What Is EBS?
Amazon Elastic Block Store (Amazon EBS) is a block storage system based in the cloud. EBS is usually used for storing data from Elastic Compute Cloud (EC2), which is Amazon’s cloud-based Virtual Machine (VM) service.
Block storage systems save huge volumes of data into blocks. Each block serves as an individual hard drive that can store any type of data. The block volumes are controlled by server-based operating systems. Due to their ability to provide high performance, block storage systems are ideal for transactional and frequently changing data.
You can deploy a variety of data types on Amazon EBS, including applications, containers, big data engines, databases, and file systems. To prevent data loss, EBS volumes are replicated in the Availability Zone (AZ). You can use EBS snapshots to backup recent changes. EBS is flexible, allowing users to change volume types and sizes, and optimize the performance.
The following features enable the management of block volumes:
- Amazon EBS Elastic Volumes━enables volumes optimization. This feature is manual, allowing you to modify the capacity and performance of the volumes, according to your needs at any certain point of time. If you want to automate this process, integrate with Amazon CloudWatch with AWS Lambda.
- Amazon EBS-Optimized instances━enables the optimization of EC2 instances for use in EBS. If you want to give your EC2 instances the same level of performance as the EBS volumes, you can use this feature, at an additional cost.
The following features enable backup for EBS volumes:
- Amazon EBS Snapshots━enables incremental backups. Snapshots backup recent changes that were made. You can use snapshots to restore previous versions of volume blocks, create new volumes by duplicating a snapshot, increase the size of an existing volume, and share a snapshot with a colleague.
- Amazon data lifecycle manager for EBS snapshots━enables automated back up. The only thing that requires configuration is the lifecycle policy, which tells the system how to create and manage backups. Then you run the policy and let the lifecycle manager do the work.
The following features secure your EBS volumes:
- Amazon EBS encryption━enables encryption of volumes and snapshots. You can use Amazon-managed keys or create your own keys in the AWS Key Management Service (KMS). The encryption system is configured to encrypt data-at-rest within the EBS system and data instances that move from and into EC2.
- AWS Identity and Access Management (IAM)━enables EBS access control. The EBS system is integrated with the AWS IAM, which provide you with controls for managing the roles and privileges of any user with access to your EBS account.
EBS Volumes types
Amazon EBS offers the use of four types of volumes, separated into two categories of physical disk drives. If you’re looking for EBS Magnetic, which used to be the standard volume type, note that it’s no longer in use.
Storage Category #1━Solid State Drives (SSD)
Solid State Drives (SSD) backed volumes are ideal for transactional workloads that need to perform many small reads and writes, measured according to their Input/Output Operations Per Second (IOPS).
1. EBS General Purpose SSD (GP2)━Default
EBS is configured with General Purpose SSD (gp2) as the default volume type. It’s ideal for boot volumes, low-latency interactive apps, and dev & test operations because it balances price and performance. This data type ensures that you get the highest performance at a cost-effective price.
2. EBS Provisioned IOPS SSD (IO1)
If you need fast performance, this is the data type for you. IO1 is ideal for I/O-intensive relational, NoSQL databases, and application workloads. IO1 can deliver a consistent performance from 50 IOPS/GB to 64,000 IOPS, while providing you with up to 1,000 MB/s of throughput per volume.
Storage Category #2━Hard Disk Drives (HDD)
Hard Disk Drives (HDD) backed volumes are ideal for large sequential workloads that rely on a high level of throughput, which is measured with MiB/s.
3. Throughput Optimized HDD (ST1)
ST1 is ideal for large, sequential workloads that define performance in terms of throughput, such as big data systems, data warehouses, and log processing workloads. You can use ST1 as a low-cost magnetic storage for MapReduce, Kafka, and Hadoop clusters.
4. Cold HDD (SC1)
As the cheapest magnetic storage option, SC1 is ideal for your coldest (infrequently accessed workloads) large sequential workloads, especially file servers. SC1 runs on a burst model, which enables you to adjust the capacity at scale.
Benefits of EBS
Amazon EBS is a great solution if you’re looking for availability and durability. To prevent loss of data, each EBS volume block is replicated across multiple servers within its allocated Availability Zone. To ensure durability, EBS volumes are configured with an annual failure rate (AFR) of between 0.1%━0.2%. If the volume fails, you can recover your latest snapshot. Another plus is the built-in integration with the AWS platform━connecting to any of your AWS workloads is a matter of a few clicks.