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.
Product Specifications and Key Documentation
|SVE/ Neon / Floating Point Unit||Included|
|Max number of CPUs in cluster||Direct-connect|
|Physical addressing (PA)||48-bit|
|Memory system and external interfaces||L1 I-Cache /D-Cache||64KB|
|L2 Cache||1MB or 512kB|
|Bus interfaces||AMBA CHI|
|Interrupts||GIC interface, GICv4|
|Debug||Armv8-A (plus Armv8.2-A extensions)|
|Embedded Trace Macrocell||ETMv4.2 (instruction trace)|
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.
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.
Talk with an Expert
Reach out to our experts for answers and resources on how to leverage the V1 Platform to create highly differentiated revolutionary system solutions.