Getting Started

Neoverse reference designs provide useful resources with best practices on how to integrate a Neoverse compute subsystem within a larger SoC. These compute subsystems are targeted at addressing requirements for specific applications in the cloud-to-edge infrastructure markets including servers, edge compute nodes, networking-storage-security offloads, mezzanine cards, 5G stations and access points. 


Neoverse N1 hyperscale reference design

The Neoverse N1 hyperscale reference design details the integration of a high core count server-class SoC subsystem using the Neoverse N1 CPU, CMN-600 coherent mesh interconnect and supporting system IP.

Diagram for the Neoverse N1 hyperscale reference design.

  • 64x Neoverse N1 CPUs, each with 1MB private L2, providing 64 threads of parallel execution
  • 8x8 coherent mesh interconnect (CMN-600) configuration with 64MB of shared system level cache
  • CPUs connected directly to the mesh in dual-core configurations for lowest latency
  • 4x CCIX links for multi-socket, chiplets and accelerator attached configurations
  • 8x memory channels supporting DDR4-3200
  • GIC-600 and CoreSight SoC-400 interfacing over the mesh
  • MMU-600 system MMU and NIC-450 non-coherent interconnect integration with PCIe Gen4 support
  • Fully compliant with Arm ServerReady 1.0 specification

Neoverse N1 edge reference design

The Neoverse N1 edge reference design details the integration of a moderate core count edge server SoC subsystem with the Neoverse N1 CPU and CMN-600 mesh interconnect.

Neoverse N1 Edge Reference Design diagram 

  • 8x Neoverse N1 CPUs, each with 512kB private L2, providing 8 threads of parallel execution
  • 4x2 CMN-600 configuration with 8MB of shared system level cache
  • CPUs are connected to the mesh through quad-core clusters with 2MB shared L3
  • 4x CCIX links to support chiplet and accelerator attached configurations
  • 2x memory channels supporting DDR4-3200
  • GIC-600 generic interrupt controller and CoreSight SoC-400 debug and trace IP interfacing over the mesh
  • MMU-600 system MMU and NIC-450 non-coherent interconnect integration with PCIe Gen4 support
  • Fully compliant with Arm ServerReady 1.0 specification

Neoverse E1 edge reference design

The Neoverse E1 edge reference design details the integration of a moderate core count high data throughput SoC subsystem with the Noeverse E1 CPU and CMN-600 mesh interconnect.

Neoverse E1 Edge Reference Design diagram 

  • 16x Neoverse E1 CPUs, each with 256kB private L2, providing 32 threads of parallel execution
  • 4x2 CMN-600 mesh configuration with 8MB of shared system level cache
  • CPUs are connected to the mesh through octa-core clusters with 2MB shared L3
  • 4x CCIX links to support chiplet and accelerator attached configurations
  • 2x memory channels supporting DDR4-3200
  • GIC-600 generic interrupt controller and CoreSight SoC-400 debug and trace IP interfacing over the mesh
  • MMU-600 system MMU and NIC-450 non-coherent interconnect integration with PCIe Gen4 support
  • Fully compliant with Arm ServerReady 1.0 specification

Request information

Want to know more about system guidance? Contact Arm today.

Contact us

Get
support

Arm Support

Arm training courses and on-site system-design advisory services enable licensees to realize maximum system performance with lowest risk and fastest time-to-market.

Arm training courses  Open a support case

Community Blogs

Community Forums

Answered Where do I find presentations and photos from SC'18? 2 votes 1802 views 0 replies Started 1 years ago by John Linford Answer this
Not answered MDK, SNTP Client, Timeout in case of error?
  • Keil MDK
  • Keil Middleware
0 votes 39 views 0 replies Started 7 hours ago by Julien MÊME (COMECA) Answer this
Suggested answer AXI4 master requirements for unaligned transactions (address v/s WSTRB) 0 votes 93 views 1 replies Latest 11 hours ago by guimers8 Answer this
Suggested answer How to set BOR for a STM32L4 micro on Keil
  • STM32 L4
  • Keil
0 votes 159 views 1 replies Latest 11 hours ago by R.Kopsch Answer this
Not answered TFT/LCD graphic contoller 0 votes 101 views 0 replies Started 14 hours ago by levetop Answer this
Suggested answer My code is not working if i change the address of flash memory , where the code can be loaded and if change the address back to the base address 0x80000000 then it works. Why? 0 votes 155 views 2 replies Latest 14 hours ago by aarushi Answer this
Answered Where do I find presentations and photos from SC'18? Started 1 years ago by John Linford 0 replies 1802 views
Not answered MDK, SNTP Client, Timeout in case of error? Started 7 hours ago by Julien MÊME (COMECA) 0 replies 39 views
Suggested answer AXI4 master requirements for unaligned transactions (address v/s WSTRB) Latest 11 hours ago by guimers8 1 replies 93 views
Suggested answer How to set BOR for a STM32L4 micro on Keil Latest 11 hours ago by R.Kopsch 1 replies 159 views
Not answered TFT/LCD graphic contoller Started 14 hours ago by levetop 0 replies 101 views
Suggested answer My code is not working if i change the address of flash memory , where the code can be loaded and if change the address back to the base address 0x80000000 then it works. Why? Latest 14 hours ago by aarushi 2 replies 155 views