aws·Jun 10, 2026·3 min readCurated summary

Now available: Amazon EC2 M9g and M9gd instances powered by new AWS Graviton5 processors | Amazon Web Services

AWS has generally released Amazon EC2 M9g and M9gd instances, powered by its Graviton5 processors. They deliver higher compute, memory, networking, and storage performance while improving energy efficiency and security. M9g targets general-purpose workloads, while M9gd adds up to 11.4 TB of local NVMe SSD storage for low-latency applications. ## Graviton5’s Performance Improvements - Graviton5 provides up to: - 25% better compute performance than Graviton4 - 35% faster web applications and ML inference - 30% faster database workloads - It includes: - 192 CPU cores - Five times more L3 cache - Up to 33% lower inter-core latency - DDR5-8800 memory - PCIe Gen6 support - These features improve concurrency, memory bandwidth, and CPU-heavy workloads such as agentic AI, code generation, real-time reasoning, and multi-step orchestration. - Customer results include: - ClickHouse: 36% better performance than M8g without code changes - Honeycomb: 36% higher throughput per core - HubSpot: up to 60% shorter MySQL query duration ## Networking and Storage Enhancements - M9g and M9gd provide: - Up to 15% higher network bandwidth - Up to 20% higher EBS bandwidth on average - Up to twice the network bandwidth on the largest instance - Instance Bandwidth Configuration lets customers shift bandwidth between EBS and VPC networking by up to 25%. - This is useful for databases, query processing, logging, and other workloads with demanding I/O requirements. ## Nitro Isolation Engine - M9g and M9gd introduce the Nitro Isolation Engine, an enhancement to the AWS Nitro System. - It isolates virtual machines by controlling access to: - Memory - CPU register state - I/O devices - The component uses formal verification to mathematically prove isolation behavior rather than relying solely on testing. - AWS describes Nitro as the first formally verified cloud hypervisor. ## M9g General-Purpose Instances - M9g instances provide one vCPU per 4 GiB of memory. - Suitable workloads include: - Application servers and microservices - Web applications and containerized applications - Midsize data stores and caching fleets - Gaming servers - Large-scale Java applications - Code repositories - Agentic AI workloads ## M9gd Instances with Local NVMe Storage - M9gd adds up to 11.4 TB of local NVMe SSD storage. - It offers up to 30% higher IOPS and storage performance than Graviton4-based M8gd instances. - Recommended for: - Low-latency caches and scratch files - Data and log processing - Media processing - Batch workloads - Key-value stores - Gaming and application servers ## Availability and Migration Resources - M9g and M9gd are available in: - US East (N. Virginia) - US East (Ohio) - US West (Oregon) - Europe (Frankfurt) - Purchasing options include Savings Plans, On-Demand, Spot Instances, Dedicated Instances, and Dedicated Hosts. - AWS provides a Graviton Getting Started Guide, Graviton Savings Dashboard, and AWS Transform for migrating Java applications from x86 to Graviton. Organizations seeking better compute efficiency, higher I/O throughput, or lower-cost Arm infrastructure should evaluate M9g for general-purpose workloads and M9gd when local NVMe storage is required.

Read original(opens in new tab)

Ddr5AI AgentsAmazon Ec2Aws Graviton5+4

cloudflare·Mar 23, 2026·3 min readCurated summary

Inside Gen 13- how we built our most powerful server yet

Cloudflare’s Gen 13 server is a hardware redesign aligned with its Rust-based FL2 request-processing stack. By choosing a 192-core AMD EPYC Turin 9965, doubling memory, expanding storage and networking, and adding stronger security and accelerator support, Cloudflare targets up to twice the throughput of Gen 12 while remaining within latency limits. The design also improves efficiency, rack density, and operational simplicity. ## Gen 13 at a Glance - Uses a 2U, single-socket design. - Key specifications: - 192-core AMD EPYC 9965 processor - 768 GB of DDR5-6400 memory - Three 7.68 TB E1.S PCIe 5.0 NVMe drives - Dual 100 GbE OCP 3.0 networking - 1,300W Titanium-grade power supply - ASPEED AST2600 BMC and AST1060 hardware root of trust - Compared with Gen 12, Gen 13 provides: - Up to 2× throughput - Up to 50% better performance per watt - Up to 60% more throughput per rack at the same power budget - Twice the memory capacity, 1.5× the storage, and 4× the network bandwidth - PCIe encryption in addition to memory encryption - Better support for high-heat PCIe accelerators ## Choosing the CPU - Gen 12 used a 96-core AMD EPYC Genoa-X 9684X with: - 400W TDP - 1,152 MB of L3 cache - Cloudflare evaluated three Turin processors: - 9755: strongest per-core performance - 9845: lower socket power and fewer cores - 9965: highest core count and better efficiency - The Turin 9965 was selected with 192 cores and 384 threads, doubling Gen 12’s hardware threads. - Its L3 cache is much smaller—384 MB total, or 2 MB per core versus Gen 12’s 12 MB per core—but FL2 workloads depend less on large L3 caches than the previous FL1 stack. - FL2 scales nearly linearly with additional cores, allowing the 9965 to deliver up to 100% higher throughput. - Production testing showed the 9965 achieved the best aggregate requests per second and favorable performance per watt at its 500W TDP. - Higher compute density also means fewer servers to provision, patch, monitor, and operate. - Turin’s support for DDR5-6400, PCIe 5.0, and CXL 2.0 provides a longer upgrade and security-support runway. ## Memory Bandwidth and Capacity - Gen 13 doubles memory from 384 GB to 768 GB while retaining 4 GB per core. - All twelve memory channels are populated using one 64 GB DDR5-6400 ECC RDIMM per channel. - This configuration delivers approximately 614 GB/s of peak memory bandwidth per socket, a 33.3% increase over Gen 12. - Using identical DIMMs across all channels enables balanced interleaving, distributing memory accesses across the full memory subsystem. - The design is intended to prevent the 192-core processor from being starved of data during highly parallel workloads. Cloudflare’s central design choice was to match hardware to the characteristics of FL2 rather than preserve Gen 12’s cache-heavy strategy. For workloads that scale well across cores, the Turin 9965 and fully populated memory system offer higher throughput, better rack economics, and simpler fleet operations.

Read original(opens in new tab)

Ddr5RustAmd EpycNvme+3