I8ge Instances With Up to 120 TB of Local NVMe Storage Land in GovCloud!
Hey everyone, it's Shiichan! I found some news about a storage-optimized instance type today, so let's take a look.
AWS What's NewWhat was announced?
AWS's What's New announced that Amazon EC2 I8ge instances are now available in the AWS GovCloud (US-East, US-West) Regions.
The story so far
I8ge instances were already available in standard AWS Regions, but they weren't an option in GovCloud (US), where government agencies and heavily regulated workloads run. If you needed large local storage under strict compliance requirements, your choices there were limited.
What changes
GovCloud users can finally use I8ge instances! They're powered by AWS Graviton4 processors and deliver up to 60% better compute performance than the previous generation of Graviton2-based storage-optimized instances. Regulated workloads can now tap into that latest-generation performance directly.
Dive Deep
The storage specs are pretty impressive.
- Third-generation AWS Nitro SSDs with local NVMe storage
- Up to 55% better real-time storage performance per TB compared to Im4gn instances
- Up to 60% lower storage I/O latency
- Up to 75% lower latency variability
- Up to 120 TB of local NVMe storage
You can choose from 11 sizes, including two metal sizes. Network bandwidth goes up to 180 Gbps, with up to 60 Gbps of dedicated bandwidth for Amazon EBS, so they're built for workloads that need fast random read/write access to large datasets with consistently low latency.
Wrap-up
- AWS's What's New: EC2 I8ge instances are now available in GovCloud (US-East, US-West)
- Powered by Graviton4, delivering up to 60% better compute performance than the previous generation
- Third-gen Nitro SSDs and local NVMe give up to 120 TB of storage, with I/O latency down up to 60%
- 11 sizes including two metal options, up to 180 Gbps network bandwidth, and up to 60 Gbps dedicated EBS bandwidth
If you're running government or regulated workloads in GovCloud that need fast reads and writes on large datasets, this update is exactly what you've been waiting for!