Executive Summary
Amazon Web Services (AWS) has announced the general availability of its new C8id, M8id, and R8id Amazon EC2 instances. These instances are powered by custom Intel Xeon 6 processors and enhance the existing 8th generation C8i, M8i, and R8i families by adding up to 22.8 TB of physically attached, high-speed NVMe-based SSD storage. Designed for I/O-intensive workloads, these new offerings provide significant performance improvements, more vCPUs, and more memory than previous generations, targeting applications like databases, data analytics, and media processing.
Key Takeaways
* Product Names: Amazon EC2 C8id, M8id, and R8id instances.
* Core Feature: Adds up to 22.8 TB of local, low-latency NVMe SSD storage to the 8th generation instance families.
* Performance Gains: Compared to sixth-generation instances, they offer up to 43% higher compute performance, 3.3x more memory bandwidth, and up to 46% higher I/O performance for database workloads.
* Instance Families:
* C8id: For compute-intensive workloads (e.g., video encoding, image manipulation).
* M8id: For balanced compute and memory workloads (e.g., data logging, media processing).
* R8id: For memory-intensive workloads (e.g., large-scale SQL/NoSQL databases, AI inference).
* Increased Scale: New, larger sizes are available, scaling up to a `96xlarge` with 384 vCPUs and 3 TiB of memory. Two bare metal sizes (`metal-48xl` and `metal-96xl`) are also offered.
* Availability: Generally available in US East (N. Virginia), US East (Ohio), and US West (Oregon). R8id is also available in Europe (Frankfurt).
* Pricing Models: Available as On-Demand, Savings Plans, Spot Instances, Dedicated Instances, and Dedicated Hosts.
Strategic Importance
This launch strengthens AWS's high-performance computing portfolio by directly addressing customer needs for workloads requiring extremely low-latency access to large datasets, reducing reliance on network-attached storage. It makes AWS more competitive for demanding database, real-time analytics, and media processing applications that were previously difficult to migrate to the cloud.