A New Benchmark for Cloud Computing, HPC, and ML Performance
The Arm Neoverse V2 CPU is designed for cloud computing, high performance computing (HPC), and machine learning (ML) performance leadership. Neoverse V2 delivers up to twice the performance as Neoverse V1. And it is the first V-series CPU with Armv9 performance, power, and security enhancements.
Features and Benefits
With a comprehensively enhanced pipeline and up to double the L2 cache size (2MB per core), Neoverse V2 can deliver twice the performance of Neoverse V1 on cloud and ML applications.
With Armv9 features like memory tagging extension (MTE) and performance defined power (PDP), Neoverse V2 is the most secure and performance-efficient V-series core yet.
When paired with the new CMN-700 core mesh network, Neoverse V2 CPUs can be configured with up to 256 cores and 512MB of system-level cache (SLC). AMBA CHI C2C enables die-to-die and off-chip connectivity using UCIe, CXL, PCIe, or custom PHYs for connecting compute, memory, and accelerators.
Performance that Spans Cloud to Edge
High-Performance Computing (HPC)
HPC researchers are exploring the scale-out performance of Neoverse N1 on computational fluid dynamics, medical research, oil and gas exploration and other computationally demanding workloads.
Cloud customers are experiencing up to 40% improved price performance over comparable traditional architecture-based instances on a broad range of workloads.
Explore More Products and Tools
The highly scalable mesh is optimized for Armv9 and Armv8-A processors, multichip configurations, and CXL attached devices. It can be customized across a wide range of performance points.
Scalable Vector Extensions
Arm’s Scalable Vector Extensions (SVE) deliver vector-length agnostic performance and programming to HPC. Write once, compile once, deploy forever.
A revolution in high-performance computing, the Neoverse V1 CPU delivers leading per-core execution and vector performance with SVE.
Our first Armv9 infrastructure CPU, Neoverse N2 offers best-in-class performance, efficiency, and compute density, while maintaining the cloud-native experience.