Amazon Web Services, today announced the general availability of Amazon Elastic Compute Cloud instances, the next generation of compute-optimized instances powered by AWS-designed Graviton3 processors. New C7g instances use AWS Graviton3 processors to provide up to 25% better compute performance for compute-intensive applications than current generation C6g instances powered by AWS Graviton2 processors. The higher performance of C7g instances makes it possible for customers to run more efficiently a wide range of compute-intensive workloads—from web servers, load balancers, and batch processing to electronic design automation (EDA), high performance computing (HPC), gaming, video encoding, scientific modelling, distributed analytics, machine learning inference, and ad serving. There are no minimum commitments or upfront fees to use C7g instances, and customers pay only for the amount of compute used.
Since launching in 2020, Amazon EC2 instances powered by AWS-designed Graviton2 processors have provided customers with significant performance improvements and cost savings for a broad range of applications. Today, 48 of the top 50 Amazon EC2 customers use AWS Graviton2-based instances to deliver superior price performance to their customers. Customers like DirecTV, Discovery, Epic Games, Formula
To provide even better price performance for a wide variety of customer applications, new C7g instances powered by next generation AWS Graviton3 processors provide up to 25% better performance for compute-intensive applications over current generation C6g instances. Compared to previous generation AWS Graviton2 processors, AWS Graviton3 processors deliver up to 2x faster performance for cryptographic workloads, up to 3x faster performance for machine learning inference, and nearly 2x higher floating point performance for scientific, machine learning, and media encoding workloads. AWS Graviton3 processors are also more energy efficient, using up to 60% less energy for the same performance than comparable EC2 instances. C7g instances are the first in the cloud to feature the latest DDR5 memory, which provides 50% higher memory bandwidth than AWS Graviton2-based instances to improve the performance of memory-intensive scientific applications like computational fluid dynamics, geoscientific simulations, and seismic processing. C7g instances also deliver 20% higher networking bandwidth than C6g instances for network intensive applications like network load balancing and data analytics.
“Customers of all sizes are seeing significant performance gains and cost savings using AWS Graviton-based instances,” said David Brown, Vice President of Amazon EC2 at AWS. “Since we own the end-to-end chip development process, we’re able to innovate and deliver new instances to customers faster. With up to 25% better performance than current generation Graviton instances, new C7g instances powered by AWS Graviton3 processors make it easy for organizations to get the most value from running their infrastructure on AWS.”
New C7g instances are built on the AWS Nitro System, a collection of AWS-designed hardware and software innovations that streamline the delivery of isolated multi-tenancy, private networking, and fast local storage. The AWS Nitro System offloads the CPU virtualization, storage, and networking functions to dedicated hardware and software, delivering performance that is nearly indistinguishable from bare metal. For customers looking to enhance the performance of applications that require parallel processing like HPC and video encoding, C7g instances in the coming weeks will include support for Elastic Fabric Adapter (EFA), which allows applications to communicate directly with network interface cards, providing lower and more consistent latency. C7g instances are available for purchase as On-Demand Instances, with Savings Plans, as Reserved Instances, or as Spot Instances. C7g instances are available today in US East (N. Virginia) and US West (Oregon), with availability in additional AWS Regions coming later this year.
Snap Inc. is a camera company focused on empowering people to express themselves, live in the moment, learn about the world, and have fun together. “We trailed the new AWS Graviton3-based Amazon EC2 C7g instances and found that they provide significant performance improvements on real workloads compared to previous generation C6g instances,” said Aaron Sheldon, Software Engineer at Snap. “We are excited to migrate our Graviton2-based workloads to Graviton3, including the messaging, storage and the friend graph workloads.”
”processors, we decided to test its performance on AWS Graviton3-based Amazon EC2 C7g instances. Rebuilding our software to run on C7g instances took us one working day, and the results were promising. When running on C7g instances, we saw 30% improved performance over comparable x86-based instances. Based on these results, we plan to recommend to our customers to run the Beamr JPEGmini software on Graviton3-based instances once those become GA, and we plan to benchmark Beamr’s H.264 and HEVC video encoders on Graviton instances as well.”
