Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage News AWS Releases New Graviton3-Based General Purpose (m7g) and Memory-Optimized (r7g) EC2 Instances

AWS Releases New Graviton3-Based General Purpose (m7g) and Memory-Optimized (r7g) EC2 Instances

Amazon Web Services (AWS) has announced the release of new Graviton3-based General Purpose (m7g) and Memory-Optimized (r7g) Amazon EC2 instances, providing customers with enhanced performance and cost savings.

The release of m7g and r7g follows the earlier release of C7g instances that were the first EC2 instances running on Graviton3 processors, which provide up to 25% better compute performance, up to 2x higher floating-point performance, and up to 2x faster cryptographic workload performance compared to AWS Graviton2 processors. 

Graviton3 brings more benefits than Graviton2 with the support of DDR5 memory, which provides 50% more bandwidth than DDR4. In addition, AWS Graviton3 processors in m7g and r7g deliver up to 25% better performance than the equivalent sixth-generation (m6g and r6g) instances, and carbon emissions are reduced.


According to AWS, the m7g and r7g instances built on the AWS Nitro System offer up to 64 vCPUs, up to 512 GB of memory, and up to 30 Gbps of network bandwidth, making them suitable for a wide range of workloads. The m7g instances are designed for general-purpose workloads, such as application servers, gaming servers, and microservices. At the same time, the r7g instances are optimized for memory-intensive workloads, such as in-memory databases and real-time big data analytics.

A respondent in a Reddit thread wrote:

The main advantage of Graviton is that you get a core per vCPU instead of a hyper thread and that they cost almost half as much overall. So yeah, you're seeing an advantage if you need many cores for CPU-bound processes. For instance, now you can use r6g.xlarge or r7g.xlarge instead of m5.2xlarge.

Public cloud providers are increasingly using arm-based CPUs, and some of the significant providers include Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform. For example, last year, Microsoft announced instances based on Ampere Altra processors. Furthermore, Google offers instances based on the Arm architecture of Compute Engine called Tau T2A (also Ampere Altra Arm-based processors). Steve Brazier, president and CEO of market research firm Canalys, believes that by 2026, some 50 percent of CPUs sold to the public clouds will be Arm-based.

M7g and R7g instances are currently available in the US East (N. Virginia), US East (Ohio), US West (Oregon), and Europe (Ireland) AWS Regions in On-Demand, Spot, Reserved Instance, and Savings Plan form.

Lastly, AWS recommends the AWS Graviton Ready Program for customers looking to move their applications to graviton instances and other resources such as the Porting Advisor for Graviton and the Graviton Fast Start program.

About the Author

Rate this Article