INFRA
INFRA
INFRA
Amazon Web Services Inc. today debuted Graviton5, the company’s next-generation custom silicon designed to deliver superior compute performance for diverse cloud workloads.
According to the company, the Graviton5 central processing unit offers up to 25% better compute performance than its predecessor, while maintaining high energy efficiency to reduce costs and meet sustainability goals.
The new chip, introduced at the AWS re:Invent conference in Las Vegas, includes a five-times larger L3 cache, a high-speed memory buffer designed to keep frequently accessed data close to the processor. This allows each Graviton5 core to access 2.6 times more L3 cache than Graviton4, which launched in July 2024. Memory performance also improved, with the new chip reducing delays waiting for data, which speeds up application response times and allows companies to run memory-intensive processes more efficiently.
Network and overall storage have also increased, with up to 15% higher network bandwidth and 20% higher Amazon Elastic Block Store bandwidth on average. This can lead to up to twice the network bandwidth for the largest instances, providing faster data transfers, quicker backups and improved performance for distributed applications.
Graviton5-based Elastic Compute Cloud M9g instances provide the largest CPU core density available in the company’s offerings, with 192 cores in a single package.
Graviton instances combine sixth-generation AWS Nitro Cards to offload virtualization, storage and networking to dedicated hardware. Nitro is the company’s security and performance foundation designed for privacy-conscious organizations, such as governments, healthcare and other regulated industries. By offloading these needs, Nitro Cards allow the CPU to run more efficiently while maintaining security.
With Graviton5, AWS introduced the Nitro Isolation Engine, enhancing the Nitro system by using formal verification to ensure mathematical certainty that workloads are isolated from each other and AWS operators.
Nitro Engine uses a minimal, formally verified codebase that consists of mathematical proofs to ensure it behaves exactly as defined. Amazon said this technology delivers a new standard for mathematically proven cloud security.
Customers will have access to the new Nitro Isolation Engine to evaluate it and the resulting proofs.
Numerous Amazon customers already use Graviton instances for cloud compute, including Adobe Inc., Airbnb Inc., Epic Games Inc., Formula One Group, Pinterest Inc. and SAP SE.
“AWS Graviton5-based Amazon EC2 instances are some of the fastest EC2 instances we have ever tested,” said Denis Sheahan, principal performance engineer at Airbnb.
Sheahan added that Airbnb’s production search workloads saw performance increases of up to 25% over other system architectures of the same generation and up to 20% compared to prior-generation Graviton4 instances.
As a global leader in software development and work management, Atlassian Corp. uses AWS Graviton instances for its Jira and Confluence services at enterprise productivity scales.
“In our testing of Jira on AWS Graviton5-based M9g instances, we observed 30% higher performance and 20% lower latency compared to the prior generation,” said Tibo Delor, principal engineer at Atlassian.
Graviton5-based M9g instances designed for general-purpose cloud workloads are available today in preview. C9g instances for compute-intensive workloads and R9g instances for memory-intensive workloads are planned for launch in 2026.
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.