Graviton5’s improved design increases speed and energy efficiency — beyond Moore’s law

Summary

Amazon announced the general availability of M9g and M9gd EC2 instances powered by the new Graviton5 processor. Built on a three-nanometer process, Graviton5 features 192 Neoverse V3 cores, DDR5-8800 memory support, PCIe gen6 interconnects, and 192 megabytes of level-three cache. The chip transitions from seven discrete dies to four unified chiplets connected at 420 gigabytes per second, eliminating dedicated I/O and memory controller dies. These architecture updates deliver up to 25% higher computational performance over Graviton4, with up to 35% faster performance for web applications and machine learning inference. The new instances also introduce the Nitro Isolation Engine, which uses formal verification to enforce hardware and software isolation between virtual machines.

Context

AWS designed Graviton5 to improve computational performance, memory bandwidth, latency, and hardware-level isolation for complex, real-world cloud workloads such as databases, web applications, and machine learning inference.

Approach / What changed

Amazon constructed Graviton5 on a three-nanometer process using 192 Neoverse V3 cores split across four chiplets, replacing dedicated I/O and DRAM dies with integrated chiplets linked by 420 GB/s interconnects. The design incorporates DDR5-8800 support, 192 MB of L3 cache, and the Nitro Isolation Engine, which applies formal verification to prove VM isolation.

Takeaways

• Graviton5 doubles core count to 192 cores across four chiplets on a 3nm process while supporting DDR5-8800 memory and PCIe gen6 interconnects.
• Replacing discrete I/O and DRAM controller dies with four integrated 48-core chiplets connected at 420 GB/s enables configurable NUMA regions and reduced memory latency for smaller VMs.
• The Nitro Isolation Engine in M9g and M9gd instances applies formal verification to mathematically prove VM isolation across memory, CPU registers, and I/O devices.

Related reading

• AWS · Jul 17, 2026

  Eclipse Dataspace Components on AWS: Cost optimization strategies

  Deploying Eclipse Dataspace Components (EDC) connectors on AWS presents challenges in forecasting infrastructure expenses without clear benchmarks. Estimated costs differ substantially between business-critical setups and non-critical environments based on compute and database sizing. A business-critical configuration relying on Amazon Aurora PostgreSQL db.r6g.large and standard Amazon ECS with AWS Fargate totals an estimated 387.00 USD monthly in the eu-south-2 region. Sizing down to db.t4g.medium and switching to AWS Fargate Spot lowers the non-critical monthly expense to 164.00 USD, representing a 58% reduction. Core cost drivers remain baseline infrastructure components, while storage, API calls, and data transfers contribute marginally under baseline assumptions.

  Jorge Hernández Suárez

• NVIDIA Developer Blog · Jun 16, 2026

  NVIDIA Blackwell Tops MLPerf Training 6.0 with Industry-Leading Scale and Performance