Amazon Web Services has released the R9g and R9gd EC2 instance families featuring the latest Graviton5 processors. These instances offer a significant performance boost, better energy efficiency, and improved hardware isolation, targeting memory-intensive workloads and demanding cloud applications.
- Up to 25% better compute performance and energy efficiency versus previous gen
- R9gd includes local NVMe SSD for low-latency storage needs
- Enhanced isolation via Nitro Isolation Engine with formal verification
Infrastructure signal
The introduction of the R9g and R9gd EC2 instances powered by AWS’s Graviton5 processors marks a key evolution in cloud infrastructure focused on memory-optimized compute capabilities. These instances deliver up to 25% improved compute performance over the earlier R8g generation, driven by faster memory speeds, higher network throughput, and a larger L3 cache. This performance gain is combined with improved energy efficiency, aligning with sustainability goals while supporting intensive workloads.
A notable infrastructure innovation is the incorporation of the Nitro Isolation Engine, which leverages formal verification to mathematically prove hardware-enforced isolation between virtual machines. This advancement raises the bar for cloud security by ensuring near bare-metal performance along with rigorous protection, reducing risks associated with multi-tenant virtualization environments. Additionally, R9gd variants introduce local NVMe SSD storage, facilitating ultra-low latency demands for databases and caching layers.
Developer impact
Developers running memory-intensive applications such as large databases, in-memory caches like Redis or Memcached, real-time analytics, and containerized microservices stand to benefit from these EC2 enhancements without needing code rewrites. The R9g family supports a multitude of popular programming languages and Linux distributions, enabling seamless migration from R8g instances with immediate performance and throughput benefits.
Container orchestration platforms including Kubernetes (EKS), Docker (ECS), and standard Linux container runtimes are fully supported, along with multi-architecture container images targeting Arm64 processors. The availability of the AWS Graviton Getting Started Guide and AWS Transform for Java further smooths adoption by simplifying workload optimization and automated code transformations for Arm-based instances.
What teams should watch
Infrastructure and cloud engineering teams should evaluate workload suitability for R9g and R9gd instances, focusing on high-memory and latency-sensitive applications to leverage the new memory bandwidth, network, and storage capabilities. Monitoring and observability strategies may need adjustment due to differences in I/O characteristics and network bandwidth allocations enabled by Instance Bandwidth Configuration (IBC).
Security teams should assess the impact of the Nitro Isolation Engine’s formal verification on compliance and risk profiles, as this provides stronger cryptographic assurance of instance isolation. Database and analytics platform teams can also explore opportunities to reduce cloud operating costs and improve resilience by shifting to these more efficient processors and taking advantage of the local NVMe storage in R9gd for caching and temporary storage needs.