Amazon EC2 I8g Instances

Best performance in Amazon EC2 for storage-intensive workloads
Get started with I8g Instances

Why Amazon EC2 I8g Instances?

Amazon Elastic Compute Cloud (EC2) I8g instances are next-generation, storage-optimized instances designed to run applications that require high throughput and real-time latency access to data on local SSD storage. I8g instances are powered by AWS Graviton4 processors and provide up to 22.5 TB of local NVMe storage using 3rd generation AWS Nitro SSDs. AWS Nitro SSDs are custom built by AWS and offer performance at cloud scale and security with always-on encryption.

I8g instances are optimized for I/O intensive workloads that require real-time latency access to data such as relational databases (MySQL, and PostgreSQL), real-time databases, NoSQL databases (Aerospike, Apache Druid, Clickhouse, MongoDB), and real-time analytics such as Apache Spark. They deliver the best compute and storage performance among storage-optimized Amazon EC2 instances.

Benefits

I8g instances based on 3rd generation AWS Nitro SSD offer up to 22.5 TB of instance storage. They provide up to 65% higher real-time storage performance per TB, 50% lower storage I/O latency, and 60% lower storage I/O latency variability compared to I4g instances.

I8g instances deliver up to 60% better compute performance compared to previous generation I4g instances. These instances include DDR5-5600 memory and are ideal for I/O intensive workloads, such as real-time databases and real-time analytics.

AWS Graviton-based instances are supported by the most popular Linux operating systems. Popular applications and services for security, monitoring and management, containers, and continuous integration and delivery (CI/CD) from AWS and partners support Graviton based instances. Customers can also find certified solutions for Graviton using the AWS Graviton Ready program.

Features

AWS Graviton4 is the latest generation of server processors designed by AWS that provide the best performance and energy efficiency for workloads in Amazon EC2. AWS Graviton4 processors deliver up to 30% better compute performance than Graviton3 processors and 60% better compute performance than Graviton2 processors.

AWS Graviton4 processors offer enhanced security with always-on memory encryption, dedicated cache for every vCPU, and support for pointer authentication. I8g instances also support encrypted Amazon Elastic Block Store.

The AWS Nitro System is a rich collection of building blocks that offloads many of the traditional virtualization functions to dedicated hardware and software. It delivers high performance, high availability, and high security, reducing virtualization overhead.

Product details

I8g instances are powered by Arm-based AWS Graviton4 processors and 3rd generation AWS Nitro SSDs. They deliver the best performance in Amazon EC2 for I/O intensive workloads.

Instance Size vCPU Memory (GiB) Instance Storage (GB) Network Bandwidth (Gbps) EBS Bandwidth (Gbps)
i8g.large 2 16 1 x 468 GB = 468 GB Up to 10 Up to 10
i8g.xlarge 4 32 1 x 937 GB = 937 GB Up to 10 Up to 10
i8g.2xlarge 8 64 1 x 1,875 GB = 1,875 GB Up to 12 Up to 10
i8g.4xlarge 16 128 1 x 3,750 GB = 3,750 GB Up to 25 Up to 10
i8g.8xlarge 32 256 2 x 3,750 GB = 7,500 GB Up to 25 10
i8g.12xlarge 48 384 3 x 3,750 GB = 11,250 GB Up to 28.125 15
i8g.16xlarge 64 512 4 x 3,750 GB = 15,000 GB Up to 37.5 20
i8g.24xlarge 96 768 6 x 3,750 GB = 22,500 GB Up to 56.25 30
i8g.metal-24xl 96 768 6 x 3,750 GB = 22,500 GB Up to 56.25 30

Customer testimonials

Here is an example of how customers and partners have achieved their business agility, price performance, cost savings, and sustainability goals with Amazon EC2 I8g instances.

  • HubSpot

    HubSpot uses Elasticsearch clusters running on Amazon EC2 instances to handle hundreds of millions of search and index requests daily. We are transitioning to OpenSearch and migrating our infrastructure to AWS Graviton-based Amazon EC2 instances for better cost efficiency. In our performance tests, the new cluster setup with Amazon EC2 I8g instances showed significant improvements over previous generation instances: up to 20% lower CPU usage, 20% better p99 search and index latencies, 40% better bulk index latencies, and 30% better delete latencies. These results translate to 100ms faster client search response times compared to the previous generation instances.

    HubSpot