AWS Launches C8id, M8id, and R8id Instances with Massive 22.8TB Local NVMe Storage
#Cloud

AWS Launches C8id, M8id, and R8id Instances with Massive 22.8TB Local NVMe Storage

Serverless Reporter
3 min read

Amazon EC2 C8id, M8id, and R8id instances powered by Intel Xeon 6 processors are now generally available, offering up to 384 vCPUs, 3TiB memory, and 22.8TB local NVMe storage for compute-intensive, memory-intensive, and balanced workloads.

AWS has announced the general availability of its new Amazon EC2 C8id, M8id, and R8id instances, delivering unprecedented local storage capacity and performance for demanding workloads. These instances, powered by custom Intel Xeon 6 processors with sustained all-core 3.9 GHz turbo frequency, represent a significant leap forward in cloud computing capabilities.

Massive Storage and Performance Improvements

The standout feature of these new instances is the up to 22.8TB of NVMe-based SSD block-level instance storage physically connected to the host server. This represents a threefold increase in vCPUs, memory, and local storage compared to previous sixth-generation instances.

Performance benchmarks show impressive gains:

  • Up to 43% higher compute performance
  • 3.3 times more memory bandwidth
  • Up to 46% higher performance for I/O intensive database workloads
  • Up to 30% faster query results for I/O intensive real-time data analytics

Instance Family Specializations

C8id instances are optimized for compute-intensive workloads requiring high-speed, low-latency local storage. These are ideal for video encoding, image manipulation, and other media processing tasks that demand rapid data access.

M8id instances strike a balance between compute and memory resources while providing high-speed local block storage. They excel at data logging, media processing, and medium-sized data stores where neither compute nor memory is the primary bottleneck.

R8id instances are designed for memory-intensive workloads including large-scale SQL and NoSQL databases, in-memory databases, large-scale data analytics, and AI inference tasks that require substantial memory bandwidth.

Scalable Configurations

These instances now scale up to 96xlarge (versus 32xlarge in the sixth generation), offering:

  • Up to 384 vCPUs
  • 3TiB of memory
  • 22.8TB of local storage

Two bare metal sizes are also available:

  • metal-48xl: 192 vCPUs, 384/768/1536 GiB memory, 11.4TB storage
  • metal-96xl: 384 vCPUs, 768/1536/3072 GiB memory, 22.8TB storage

Flexible Resource Allocation

The instances support the Instance Bandwidth Configuration (IBC) feature, allowing customers to allocate resources between network and Amazon Elastic Block Store (Amazon EBS) bandwidth with 25% scaling flexibility. This enables optimal resource allocation based on specific workload requirements.

Enhanced Security and Performance

These instances utilize sixth-generation AWS Nitro cards, offloading CPU virtualization, storage, and networking functions to dedicated hardware and software. This architecture enhances both performance and security for workloads.

Local NVMe devices feature hardware encryption using the XTS-AES-256 block cipher with unique keys that are destroyed when instances stop or terminate. The storage has the same lifetime as the instance and does not persist after termination.

Regional Availability and Pricing

C8id, M8id, and R8id instances are available in US East (N. Virginia), US East (Ohio), and US West (Oregon) regions. R8id instances are additionally available in the Europe (Frankfurt) region.

Customers can purchase these instances as:

  • On-Demand Instances
  • Savings Plans
  • Spot Instances
  • Dedicated Instances
  • Dedicated Hosts

For detailed pricing information, visit the Amazon EC2 Pricing page.

Getting Started

The instances are available in 11 sizes per family plus two bare metal configurations. To begin using these powerful new instances, visit the EC2 console or check the EC2 C8i instances, M8i instances, and R8i instances page for more information.

These instances represent AWS's continued investment in providing customers with high-performance computing options that can handle the most demanding workloads while offering flexibility in resource allocation and pricing models.

Comments

Loading comments...