Pace of Innovation for Custom Silicon on Arm Continues with AWS Graviton4
AWS announces a new Arm Neoverse-based CPU, the AWS Graviton4.
It’s been said that the best way to predict the future is to create it. This anecdote is clearly reflected in the data center and infrastructure market we see today being built on Arm. Ten years ago, no one had heard of a data processing unit (DPU); and it was early days for the Arm CPU server market. But some insightful and innovative people at Amazon had the vision to see where compute was heading, and that vision led to investments and partnerships and to what we now know today as AWS Nitro – one of the first true DPUs – and AWS Graviton – the first CPU custom-built by a hyperscale cloud provider.
Both represented the future of purpose-built infrastructure for the era of AI and workload-optimized computing we see clearly emerging at scale today. Having worked closely with AWS for almost 10 years, it’s fantastic to see what they have brought to market. This collaboration between AWS and Arm is at the forefront of driving the transformational changes we are seeing in the datacenter.
Today, we celebrate another major milestone as AWS announces a new Neoverse-based CPU, the AWS Graviton4. This marks the fourth Graviton CPU in five years – an incredible pace of innovation, and proof of the advantages of building custom silicon on Arm. Graviton4 supports 50 percent more cores and up to 75 percent more memory bandwidth per CPU compared to its predecessor, AWS Graviton3. These capability upgrades result in some very impressive performance improvements – up to 30 percent faster for web applications, 40 percent for databases, and 45 percent for large Java applications.
According to AWS, Graviton4 is the most powerful and energy efficient chip they have ever built. Best of all? Across their entire fleet, Graviton4 delivers the best performance and energy efficiency for a broad range of workloads running on Amazon EC2.
To read the full article, click here
Related Semiconductor IP
- eDP 2.0 Verification IP
- Gen#2 of 64-bit RISC-V core with out-of-order pipeline based complex
- LLM AI IP Core
- Post-Quantum Digital Signature IP Core
- Compact Embedded RISC-V Processor
Related Blogs
- Reduce H.265 High-Res Encoding Costs by over 80% with AWS Graviton2
- System Verification of Arm Neoverse V2-Based SoCs
- EDA in the Cloud: Astera Labs, AWS, Arm, and Cadence Report
- Neoverse CSS V3: TCO-Optimized Confidential Compute for Cloud
Latest Blogs
- Enhancing PCIe6.0 Performance: Flit Sequence Numbers and Selective NAK Explained
- Smarter ASICs and SoCs: Unlocking Real-World Connectivity with eFPGA and Data Converters
- RISC-V Takes First Step Toward International Standardization as ISO/IEC JTC1 Grants PAS Submitter Status
- Running Optimized PyTorch Models on Cadence DSPs with ExecuTorch
- PCIe 6.x: Synopsys IP Selected as First Gold System for Compliance Testing