Amazon EC2 R7iz instances
The fastest 4th Generation Intel Xeon Scalable-based instances in the cloud
Why Amazon EC2 R7iz Instances?
Amazon Elastic Compute Cloud (EC2) R7iz instances are memory-optimized, high CPU performance instances. They are the fastest 4th Generation Intel Xeon Scalable-based (Sapphire Rapids) instances in the cloud with 3.9 GHz sustained all-core turbo frequency. R7iz instances deliver up to 20% better performance than previous generation z1d instances. They use DDR5 memory and deliver up to 2.4x higher memory bandwidth than z1d instances. R7iz instances feature an 8:1 ratio of memory to vCPU with up to 128 vCPUs and up to 1,024 GiB of memory. The combination of high CPU performance and high memory footprint makes R7iz instances ideal for frontend Electronic Design Automation (EDA), relational database workloads with high per-core licensing fees, and financial, actuarial, and data analytics simulation workloads.
AWS and Intel Partnership
AWS and Intel continue to collaborate to provide cloud services that are designed to meet current and future computing requirements. For more information, see the AWS and Intel partner page.
Benefits
Lower TCO
R7iz instances deliver up to 20% higher compute performance than previous generation z1d instances. High CPU performance combined with up to 1024 GiB memory results in increased overall performance for applications such as EDA and relational databases. This can help you reduce time to market for product development while reducing licensing costs.
Flexibility and choice
R7iz instances add to the broadest and deepest selection of EC2 instances in the cloud. They provide instance sizes up to 32xlarge and offer two bare metal sizes. With up to 2.6x more vCPUs than other high-frequency instances, you can scale up your memory-intensive workloads.
Maximized resource efficiency
R7iz instances are built on the AWS Nitro System, a combination of dedicated hardware and lightweight hypervisor that delivers practically all of the compute and memory resources of the host hardware to your instances for better overall performance and security.
Features
High CPU performance
R7iz instances are the fastest 4th Generation Intel Xeon Scalable-based (Sapphire Rapids) instances in the cloud with 3.9 GHz sustained all-core turbo frequency. These instances include support for always-on memory encryption using Intel Total Memory Encryption (TME).
High-performance interfaces
R7iz instances support up to 50 Gbps networking. R7iz instances also support 40 Gbps bandwidth to Amazon Elastic Block Store (EBS) in the largest size. Additionally, with R7iz instances, you can attach up to 88 EBS volumes to an instance (compared to z1d which allowed up to 28 EBS volume attachments to an instance). R7iz instances use the new DDR5 memory technology and provide up to 2.4x higher memory bandwidth than comparable high-frequency instances. R7iz instances have support for Elastic Fabric Adapter (EFA) in the 32xlarge and the metal-32xl sizes.
New accelerators
4th Gen Intel Xeon Scalable processors offer 4 new built-in accelerators. Advance Matrix Extensions (AMX) – available on all sizes - accelerate matrix multiplication operations for applications such as CPU-based machine learning. Data Streaming Accelerator (DSA), In-Memory Analytics Accelerator (IAA), and QuickAssist Technology (QAT) - available on R7iz bare metal sizes - enable efficient offload and acceleration of data operations that help in optimizing performance for databases, encryption and compression, and queue management workloads.
Built on the Nitro System
The AWS Nitro System can be assembled in many different ways, allowing AWS to flexibly design and rapidly deliver EC2 instance types with an ever-broadening selection of compute, storage, memory, and networking options. Nitro Cards offload and accelerate I/O for functions, increasing overall system performance.
R7iz instances
Amazon EC2 R7iz instances are powered by 4th Generation Intel Xeon Scalable processors and are an ideal fit for high CPU and memory-intensive workloads.
Instance Size
|
vCPU
|
Memory (GiB)
|
Instance Storage (GB)
|
Network Bandwidth (Gbps)
|
EBS Bandwidth (Gbps)
|
---|---|---|---|---|---|
r7iz.large
|
2 |
16 |
EBS-Only |
Up to 12.5 |
Up to 10 |
r7iz.xlarge
|
4 |
32 |
EBS-Only |
Up to 12.5 |
Up to 10 |
r7iz.2xlarge
|
8 |
64 |
EBS-Only |
Up to 12.5 |
Up to 10 |
r7iz.4xlarge
|
16 |
128 |
EBS-Only |
Up to 12.5 |
Up to 10 |
r7iz.8xlarge
|
32 |
256 |
EBS-Only |
12.5 |
10 |
r7iz.12xlarge
|
48 |
384 |
EBS-Only |
25 |
19 |
r7iz.16xlarge
|
64 |
512 |
EBS-Only |
25 |
20 |
r7iz.32xlarge
|
128 |
1,024 |
EBS-Only |
50 |
40 |
r7iz.metal-16xl
|
64 |
512 |
EBS-Only |
25 |
20 |
r7iz.metal-32xl
|
128 |
1,024 |
EBS-Only |
50 |
40 |
Customer testimonials
Here are examples of how customers and partners have achieved their business agility, price performance, cost savings, and sustainability goals with Amazon EC2 R7iz instances.
Aiven

Astera Labs

Nasdaq

Akami

SingleStoreDB
SingleStoreDB is a cloud-native database built for speed and scale to power real-time applications.
" Leading companies across nearly every vertical around the world use SingleStoreDB to enhance customer experience and to improve operations and security. Optimizing the compute performance of the underlying infrastructure is necessary to support constantly growing workloads. When testing Amazon EC2 R7iz instances, our engineering teams saw a 19% improvement on database performance versus prior generation Ice Lake based instances. We look forward to leveraging the Amazon EC2’s latest Sapphire Rapids instances to deliver exceptional performance for transactions and analytics. "
Rob Weidner, Director of Cloud Partnerships, SingleStoreDB

TotalCAE
TotalCAE's platform supports hundreds of engineering applications and makes it simple for customers to adopt High Performance Computing (HPC) applications in the cloud.
" Amazon EC2 R7iz instances combine 1 TB of the newest DDR5 memory and the latest 4th Generation Intel Xeon Scalable processors running at 3.9 GHz to offer next generation performance for applications such as Finite Element Analysis (FEA). We tested several flagship licensed FEA applications on R7iz instances and observed performance gains of up to 19% for the same license cost over previous generation R6id instances. Our clients invest heavily in their FEA application licenses, and we are eager to help them maximize their license investments and accelerate their time to market. "
Rod Mach, President, TotalCAE

Amazon Relational Database Service (RDS)
Amazon Relational Database Service (RDS) is a collection of managed services that makes it simple to set up, operate, and scale databases in the cloud.
" Amazon EC2 R7iz instances are ideal for relational database workloads that typically have high per-core licensing costs. Our Airline and Banking customers running demanding workloads currently use z1d instances. R7iz's 20% higher compute performance, larger sizes (up to 32xlarge), and 2.4x memory throughput (using the latest DDR5) versus z1d will help these customers achieve superior performance as they continue to scale. "
Kambiz Aghili, GM, RDS, DBS Managed Commercial Engines, AWS

Hugging Face
The Hugging Face Hub works as a central place where anyone can share, explore, discover, and experiment with open-source ML.
"At Hugging Face, we are proud of our work with Intel to accelerate the latest models on the latest generation of hardware, from Intel Xeon CPUs to Habana Gaudi AI accelerators, for the millions of AI builders using Hugging Face.
The new acceleration capabilities of Intel Xeon 4th Gen, readily available on Amazon EC2, introduce bfloat16 and INT8 support for transformers training and inference, thanks to Advanced Matrix Extensions (AMX).
By integrating Intel Extension for Pytorch (IPEX) into our Optimum-Intel library, we make it super easy for Hugging Face users to get the acceleration benefits with minimal code changes. Using the custom EC2 Gen 7 instances (such as Amazon EC2 R7iz and other instances), we reached an 8x speedup fine-tuning DistilBERT and were able to run inference 3x faster on the same transformers model. Likewise, we achieved a 6.5x speedup when generating images with a Stable Diffusion model.”
Ella Charlaix, ML Engineer, Hugging Face
