Amazon Web Services has made generally available its T8i burstable EC2 instances, which run on custom sixth-generation Intel Xeon Scalable Processors (Granite Rapids) exclusive to AWS. Positioned among the most affordable EC2 options, T8i instances deliver up to 30% better price performance compared to the earlier T3 generation. The offering targets workloads with low-to-moderate CPU demands, including freemium applications, training environments, staging and development systems, data processing, microservices, low-traffic web properties, and authentication gateways.

Use cases and customer demand

Substantial numbers of AWS customers currently operate lightweight workloads on T3 instances that benefit from small, economical compute resources. These span microservices deployments, low-traffic websites, development and testing setups, compact databases, data processing operations, and brief compute jobs. The T family's burstable performance model appeals to these users because it establishes a baseline CPU level while permitting temporary performance increases through CPU credits.

As organizations refresh their computing infrastructure, transition from on-premises systems, embrace event-driven and microservices patterns, and explore AI inference applications, they have sought newer cost-optimized small instances, enhanced price-to-performance ratios to minimize total cost of ownership, and straightforward migration options that build on current knowledge and existing tools.

Key capabilities

  • Up to 30% better price performance. Built on the AWS Nitro System with custom sixth-generation Intel Xeon Scalable Processors (Granite Rapids), T8i instances reduce total cost of ownership through superior price-to-performance metrics.
  • Up to 70% higher compute performance. T8i instances provide up to 70% increased compute capacity, up to 1.25x greater network bandwidth, and up to 2.4x faster EBS bandwidth relative to T3 instances.
  • Straightforward migration from T3. Existing T3 customers can upgrade without complexity. T8i maintains the same CPU credit mechanism and lightweight compute options T3 users already employ. Switching to T8i requires only selecting the new instance type to gain performance improvements.
  • Accessible entry point for new deployments. Organizations new to AWS or transitioning from on-premises infrastructure can leverage T8i instances as one of the most economical starting points for low-to-moderate CPU workloads or short-lived compute tasks such as batch jobs, event-triggered functions, or continuous integration and deployment pipelines.

Instance specifications

T8i instances come in four sizes, each featuring two vCPUs configured as a single core. Like T3, T8i instances feature distinctive vCPU-to-memory configurations including 1:0.25, 1:0.5, and 1:1 ratios unavailable on other EC2 instance types. Both generations employ the CPU credit system with Standard and Unlimited credit modes, with Unlimited as the default for T8i.

For workloads exceeding T8i capacity (nano, micro, small, and medium sizes), M8i Flex instances deliver up to 30% better price performance than comparable T3 predecessors while offering scalability up to 16xlarge.

Availability and purchasing

T8i instances are available immediately across multiple AWS Regions: US East (N. Virginia, Ohio), US West (Oregon, N. California), Asia Pacific (Hyderabad, Malaysia, Mumbai, Seoul, Singapore, Sydney, Tokyo), Canada (Central), and Europe (Frankfurt, Ireland, London, Paris). Regional availability and planned expansion can be found through the CloudFormation resources section of AWS Capabilities by Region.

Customers can acquire T8i instances through On-Demand purchasing, with Spot instances and Savings Plan options arriving soon. T8i instances support shared tenancy only and do not support Dedicated tenancy or Dedicated Hosts. The t8i.micro and t8i.small sizes qualify for the AWS Free Tier. Additional details are available on the Amazon EC2 Pricing page.

Users can test T8i instances through the Amazon EC2 console and provide feedback via AWS re:Post for EC2 or standard AWS Support channels.