4th Gen AMD EPYC vs Competitors Cloud Performance

4th Gen AMD EPYC Processors

Amazon EC2 M7a instances bring an even greater variety of workloads to AWS and provide up to 50% more compute performance than M6a instances do.

When compared to Amazon EC2 Hpc6a instances, 4th Gen AMD EPYC processor-powered Amazon EC2 Hpc7a instances offer up to 2.5x the performance.

The general availability of Amazon Elastic Compute Cloud (EC2) M7a and Amazon EC2 Hpc7a instances, which provide next-generation performance and efficiency for applications that benefit from high performance, high throughput, and tightly coupled HPC workloads, respectively, was announced today by AMD.

According to David Brown, vice president of Amazon EC2 at AWS, “4th Gen EPYC processor-powered Amazon EC2 instances deliver a differentiating offering for customers with increasingly complex and compute-intensive workloads.” Both M7a and Hpc7a instances enable quick and low-latency internode communications when combined with the strength of the AWS Nitro System, enhancing what our customers can accomplish across our expanding family of Amazon EC2 instances.

“4th Gen EPYC processors have unmatched performance leadership; and our collaboration with AWS is delivering the full 4th Gen EPYC performance per core in new M7a and Hpc7a instances,” said Dan McNamara, senior vice president and general manager, AMD’s Server Business Unit. As a result, “our customers can optimize their businesses while performing more work with less resources across a variety of workloads.”

EC2 is being expanded by AMD and AWS for improved performance and new applications

The Amazon EC2 M7a instances support a variety of workloads and offer customers up to 50% more compute performance than Amazon EC2 M6a instances while also introducing new processor capabilities like AVX3-512, VNNI, and BFloat16. They were first showcased at this year’s AMD Data Center and AI Technology Premiere. The Amazon EC2 M7a instances are excellent for applications that benefit from high performance and high throughput, such as financial applications, simulation modeling, gaming, mid-sized data stores, and more, in addition to providing high performance for general-purpose workloads.

When compared to Amazon EC2 Hpc6a instances, Amazon EC2 Hpc7a instances perform 2.5 times better and are intended for tightly coupled high performance computing workloads. These instances provide AWS customers with increased compute, memory, and network performance to meet the demands of escalating workload complexity. The Hpc7a instances are perfectly suited for compute-intensive, latency-sensitive workloads like computational fluid dynamics, weather forecasting, molecular dynamics, and computational chemistry thanks to their 300 Gb/s EFA network bandwidth. Next-generation server technology is also included in Amazon EC2 Hpc7a instances, such as DDR5 memory, which offers 50% more memory bandwidth than DDR4 memory to enable high-speed access to data in memory.

“DTN deploys a suite of weather data and models that deliver sophisticated, high-resolution outputs and require continual processing of vast amounts of data from inputs across the globe,” said Lars Ewe, Chief Product and Technology Officer, DTN. “To deliver high-level operational intelligence for weather-dependent industries, DTN deploys a suite of weather data and models.” “Thanks to our partnership with AWS, we are able to provide our customers with the most recent weather data, which powers those analytical engines. We are eager to learn how the upcoming Amazon EC2 Hpc7a instances may help us in fulfilling our goal of giving customers the insights they require at the precise moment they are required.

In 2018, AMD and AWS started collaborating, and as of right now, AMD offers over 100 EPYC processor-based instances for general-purpose, compute- and memory-optimized, and high-performance computing workloads. With AMD-based Amazon EC2 instances, customers like DNT, Sprinklr, and TrueCar have all benefited from significant cost and cloud utilization optimization. These new instances allow AWS users to continue benefiting from AMD EPYC processors‘ excellent performance, scalability, and efficiency.