Enabling Virtualized Telco Network Services
Virtual BBU, O-DU and O-CU Acceleration in the Edge Cloud
Telco Acceleration
5G operators are steadily moving away from traditional proprietary wireless equipment in favor of an open, disaggregated telco server approach. Whether it’s O-RAN or vBBU, this “virtualization of 5G” carries with it the promise of telco-deployed software services at the edge such as video streaming, gaming, or automotive services.
AMD Telco Accelerator cards offload the latency-sensitive and throughput intensive 5G baseband functions, freeing up telco server processors for more interesting and commercialized software functions. These cards deliver the right performance, power, and ease of deployment required for the 5G virtualized edge.
Why AMD Telco Accelerator Cards
High-Performance 5G Offload
Virtualizing the baseband only works if there’s enough processing horsepower in your server and accelerator to keep up with the rigorous demands of 5G.
 |
 |
|
| T1 Accelerator | T2 Accelerator | |
|---|---|---|
| Optimization | Frounthaul & L1 | L1 |
| Form Factor | FHHL | HHHL |
| PCIe | 2x Gen3x8 | Gen3x16 or 2x Gen4x8 |
| FPGA | ZU19P + ZU21DR | ZU48DR |
| FH Ports | 2x SFP28 + 1588 | N/A |
| L1 Encode | 17.7 Gbps | 35 Gbps |
| L1 Decode | 8.1 Gbps | 12 Gbps |
Available Cards
Different 5G deployments have differing hardware acceleration needs. AMD has you covered.
T1 Telco Accelerator Card
AMD T1 Telco Accelerator card turns a standard server into a virtual baseband unit with the performance, low latency, and power efficiency needed for O-RAN 5G deployments. T1 cards can be used to accelerate both Forward Error Correction (FEC) and Fronthaul (FH) functions of L1 Baseband processing. The flexibility to support diverse workloads makes the radio network more agile and scalable while reducing the total costs of deployment.
Deliver on the promise of O-RAN.
T1 – 5G Fronthaul Reference Design
Layer 1 Offload Reference Design
T2 Telco Accelerator Card
AMD T2 Telco Accelerator Card provides the highest Forward Error Correction performance on a PCIe gen4 card in a lookaside configuration in dense, urban deployments enabling a highly reliable and efficient radio network. The turnkey solution enables operators, system integrators, and OEMs to get to market quickly and to simplify the deployment of services at the edge.
T2 – Layer 1 Reference Design
Turnkey Options for Ease-of-Deployment
Telco Accelerator reference designs remove the traditional barriers to FPGA accelerator deployment.
Traditional FPGA accelerators require the OEM or telco to have a team of RTL and firmware designers to develop, integrate, and deploy the accelerator. AMD Telco Accelerator cards, starting with T1, remove this obstacle by being offered with a pre-developed and tested reference designs for fronthaul and layer-1 offload. The cards take advantage of open source software stacks used by today’s leading system integrators for ease of design. View demo.
The VVDN reference design and software stack are available from AMD to ease integration and performance testing of the card. Deployment licenses are commercially available from VVDN.
Featured Videos
Get Started
Find evaluation boards, leverage libraries to develop your own applications, and learn how to become a AMD wired & wireless IP partner.
Use a Pre-built Evaluation Board
Design with a leading-edge using one of our evaluation boards
Explore now >
Develop Your Own Applications
Learn about developing your own applications using Vitis™ AI
Learn more >
Partner Program
Expand your reach and solve more customer problems by partnering with us
Join now >
Stay Informed
Sign up for Wired & Wireless Updates
Inquiries
Contact Sales for More Information