A Revolution in High Performance Computing
The Neoverse V1 CPU is designed to deliver leading per-core performance on demanding HPC, HPC in the cloud and AI/ML-assisted workloads. Neoverse V1 is also the first Neoverse processor to include Scalable Vector Extension (SVE) for maximum vector performance, HPC code re-use, and longevity.
Features and Benefits
Delivers an incredible 50 percent IPC performance uplift over the Neoverse N1 core thanks to a new micro-architecture designed for maximum per-core performance.
Includes two 256b SVE vector units, doubling the integer and floating-point execution capability of Neoverse N1 per core for better HPC and AI/ML performance. SVE was invented in collaboration with Fujitsu for the Fugaku project and enables vector-width agnostic programming. SVE enables code execution of vectors ranging from 128b to 2048b wide on Neoverse V1 without requiring recompilation.
Enables the deployment of high-bandwidth DDR5 and HBM2e/3 memory systems to unlock the full performance of Neoverse V1 cores. Designers can augment core performance with custom or third-party accelerators connected over the flexible CMN-700 interconnect.
The Arm Neoverse V1 CPU ushers in a new performance tier to the Neoverse product portfolio.
Get Developer Resources for more details.
|Arm Neoverse V1 Technical Reference Manual|
|Arm Neoverse V1 Software optimization guide|
|Arm Neoverse V1 reference designs|
Performance that Spans Cloud to Edge
High Performance Computing (HPC)
HPC researchers are exploring the scale-out performance of Neoverse architectures on computational fluid dynamics, medical research, oil & gas exploration and other computationally demanding workloads.
More customers are shifting their workloads in Computational Fluid Dynamics, Genomics, Oil & Gas, and EDA to the cloud. Arm Neoverse offers excellent performance throughput and value.
Neoverse V1 is the best Arm core for accelerating AI/ML-assisted workloads, helping keep specialized computing resources fed quickly with data.
Explore More Products and Tools
Scalable Vector Extensions
Arm’s Scalable Vector Extensions (SVE) deliver vector-length agnostic performance and programming to HPC. Write once, compile once, deploy forever.
The highly scalable mesh is optimized for Armv9 and Armv8-A processors, multi-chip configurations, and CXL attached devices. It can be customized across a wide range of performance points.
Our first Armv9 infrastructure CPU, Neoverse N2 offers best-in-class performance, efficiency, and compute density, while maintaining the cloud native experience.
Experts from ETRI and SiPearl describe how Arm Neoverse V1 supports their goal of building exascale-class CPUs for AI and other HPC projects, with state-of-the-art power and performance efficiency.