virtex-ultrascale-plus

Introducing the Virtex UltraScale+ VU23P FPGA


Optimized for Adaptable Networking and Storage Acceleration
 

vu23p-chip

Overview

Optimized Resources for Adaptable Networking and Storage Acceleration

  • Highest ratio of LUTs1 and block RAM to DSP slices in the Virtex UltraScale+ portfolio
  • Maximize packet processing throughput per port with adaptability
  • Scale data bandwidth within a fixed power envelope

Unique Mix of 58G PAM4 Transceivers and PCIe Gen4

  • PCIe® Gen4 for maximum I/O bandwidth
  • Integrated 100G Ethernet MAC with KR4-FEC
  • 58G PAM4 transceivers for up to 200G SmartNIC and edge network systems

Right-Sized Packaging Options for Server Deployment

  • Package options as small as 35mmx35mm
  • Flexibility to support a broad range of standard server form factors

Solutions

The exponential growth of data demands smart network and data center solutions that reach beyond fixed-function silicon devices to deliver maximum throughput, high data processing capability, and the flexibility to adapt to evolving connectivity standards. Accelerate network and data center systems with the Virtex™ UltraScale+™ VU23P FPGA.

 

Compute Acceleration

SmartNICs: More than Just Network Processing

By offloading network intensive workloads (SDN/NFV) from host CPU to FPGA-based SmartNICs, users can significantly boost server performance, reduce system-level total cost of ownership and enable rapid feature innovation. Customers leveraging a Virtex UltraScale+ VU23P FPGA can enable up to 2x100G applications, depending on workloads. The small package options create the flexibility to build diverse solutions in 1U and 2U server platforms, with additional DDR and control plane processors when necessary. Overall, Virtex UltraScale+ VU23P FPGA-based SmartNICs enable highly programmable and performant solutions that provide outstanding agility and scalability for the data center.


Compute Acceleration

NVMe-oF: A New Way to Store and Move Data

For next-generation storage systems, FPGAs are ideal for delivering improved performance in systems with RDMA and NVMe over Fabrics (NVMe-oF) offloads, either integrated as SSD controllers or in hyperconverged infrastructure to handle complex storage processing tasks. The Virtex UltraScale+ VU23P FPGA enables increased acceleration of a variety of types of storage workloads including data compression, decompression, deduplication, and sequencer functionality while improving overall storage utilization and minimizing the burden on the CPU. With package options as small as 35x35mm and PCIe Gen4 compatibility, the Virtex UltraScale+ VU23P FPGA offers unparalleled performance per SQMM, creating flexibility in deployments and reducing total cost of ownership of storage systems.


Compute Acceleration

Converged Access Fronthaul: Adaptable Edge Networks

Converged access fronthaul gateways allow all services to be carried across a single interface without compromising performance. The Virtex UltraScale+ VU23P FPGA enables different device personalities to be selected for each deployment within the network infrastructure, creating the flexibility to support a variety of standards including fronthaul over Ethernet, fronthaul over OTN, and fronthaul over PON. Furthermore, the integrated PCIe Gen4 IP provides a path to virtualize baseband processing for existing radios, enabling virtual broadband base units (vBBUs) and LO PHY processing in the transport box. 

GET STARTED

 

1: Highest ratio of LUT to DSP slices in the Virtex UltraScale+ Portfolio, excluding the Virtex UltraScale+ VU19P FPGA