Design Flow Assistant
Interactive guide to help you create a development strategy.
Enabling High Data Throughput with the Industry’s Fastest DDR Memory Interface
Provides a diverse set of compute engines, next-generation I/O, and integrated DDR controllers, enabling low-latency acceleration across a wide range of workloads
Three scalar processor types are available in Versal adaptive SoCs to support diverse application needs. The application processing unit is ideal for complex applications supported by an OS, and the real-time processing unit is ideal for applications which need low latency, determinism, and real-time control. A separate platform management controller manages system boot, security, and debug.
The Prime series is a highly integrated, multicore, heterogeneous compute platform that delivers breakthrough performance for a wide range of applications, including data center networking, storage, and wired communications, by enabling low-latency inline acceleration in devices optimized for connectivity.
With the dawn of AI and the explosion of data generated by modern applications, traditional compute has become an increasingly precious resource in the data center. One solution to preserve compute is to add accelerators to storage controllers and endpoints to offload compression and other common tasks from the primary processor. The Versal Prime series combines PCIe® Gen5/Gen4 support, powerful Arm® scalar processing engines, integrated memory controllers, and next-generation programmable logic to reduce system-level bottlenecks and increase efficiency.
FPGAs and SoCs have long played a key role in accelerating data center applications by offloading common networking functions such as encryption/decryption and virtual switching from the central processor in a system, preserving compute cycles. Versal adaptive SoCs further accelerate data center systems by combining PCIe Gen5/Gen4 support, next-generation SerDes, and a multirate Ethernet MAC with powerful heterogeneous compute engines to both offload networking operations and supplement the compute available in a server with additional inline processing.
As 5G is emerging, the demand for high-bandwidth connectivity requires access networks to adapt to support new protocols, such as eCPRI, while still maintaining compatibility with existing standards. The combination of adaptable engines with industry-leading SerDes and Ethernet technology in the Versal Prime series gives operators the processing power needed to handle 5G throughput requirements while maintaining legacy standard support and reducing total cost of ownership.
As competition from wireless services increases, cable broadband providers must scale their networks and deliver higher bandwidth service to remain competitive. The Versal Prime series offers twice the programmable logic as existing MPSoC devices, powerful Arm scalar processing engines, greater SerDes density, integrated Ethernet IP blocks, and thermally optimized packaging to enable low power, efficient designs without compromising performance.
The massive explosion in bandwidth that accompanies the latest communications standards makes it challenging for test equipment vendors to keep pace with market demands. With support for 58G transceivers and multirate Ethernet MACs, the Versal Prime series offers high-performance data handling and analysis for wired communication testers. The programmable NoC and integrated memory controllers enable fast and efficient data movement for complex test algorithms.
Introducing the Versal Prime VM2152 device, featuring the industry’s fastest DDR memory interface1, supporting LPDDR5 (up to 6400 Mb/s) / DDR5 (up to 5600 Mb/s), 112 Gb/s transceivers, and increased DSP capability2 (more than previous Versal Prime devices) to support the increased data throughput needed in next-generation communications, test & measurements, storage, and other broad market applications.
VM1102 | VM1302 | VM1402 | VM1502 | VM1802 | VM2152 | VM2202 | VM2302 | VM2502 | VM2902 | |
---|---|---|---|---|---|---|---|---|---|---|
Application Processing Unit | Dual-core Arm® Cortex®-A72, 48 KB/32 KB L1 Cache w/ parity & ECC; 1 MB L2 Cache w/ ECC | |||||||||
Real-time Processing Unit | Dual-core Arm Cortex-R5F, 32 KB/32 KB L1 Cache, and 256 KB TCM w/ECC | |||||||||
Memory | 256 KB On-Chip Memory w/ECC | |||||||||
Connectivity | Ethernet (x2); USB 2.0 (x1); UART (x2); SPI (x2); I2C (x2); CAN-FD (x2) |
VM1102 | VM1302 | VM1402 | VM1502 | VM1802 | VM2152 | VM2202 | VM2302 | VM2502 | VM2902 | |
---|---|---|---|---|---|---|---|---|---|---|
DSP Engines | 464 | 848 | 1,696 | 1,312 | 1,968 | 1,704 | 1,312 | 1,904 |
3,984 | 2,672 |
VM1102 | VM1302 | VM1402 | VM1502 | VM1802 | VM2152 | VM2202 | VM2302 | VM2502 | VM2902 | |
---|---|---|---|---|---|---|---|---|---|---|
System Logic Cells (K) | 329 | 703 | 1,238 | 981 |
1,968 | 757 | 1,139 | 1,575 | 1,969 | 2,233 |
LUTs | 150,272 | 321,536 | 565,760 | 448,512 |
899,840 | 346,112 | 520,704 | 719,872 | 900,224 | 1,020,928 |
VM1102 | VM1302 | VM1402 | VM1502 | VM1802 | VM2152 | VM2202 | VM2302 | VM2502 | VM2902 | |
---|---|---|---|---|---|---|---|---|---|---|
NoC Master / NoC Slave Ports | 5 | 9 | 18 | 21 |
28 | 12 | 21 | 30 | 28 |
42 |
DDR4 Memory Controllers | 1 | 2 | 4 | 3 |
4 | - | 3 | 3 |
4 |
3 |
DDR5 Memory Controllers | - | - | - | - | - | 4 | - | - | - | - |
PCIe® w/ DMA (CPM) | - | 1 x Gen4x16 | 1 x Gen4x16 | 1 x Gen4x16 | 1 x Gen4x16 | - | 2 x Gen5x8 | - |
2 x Gen5x8 | - |
PCIe | 1 x Gen4x8 | 2 x Gen4x8 | 2 x Gen4x8 | 4 x Gen4x8 | 4 x Gen4x8 | 2 x Gen5x4 | 4 x Gen5x4 | 2 x Gen5x4 | - | 2 x Gen5x4 |
100G Multirate Ethernet MAC | 1 | 2 | 2 | 4 | 4 | 2 | 2 | 6 | - | 6 |
600G Ethernet MAC | - | - | - | - | - | 1 | - | - | - | - |
400G High Speed Crypto | - | - | - | - | - | 1 | - | - | - | - |
GTY Transceivers | 0 | 24 | 24 | 44 | 44 | 0 | 0 | 0 | 0 | 0 |
GTYP Transceivers | 8 |
0 | 0 | 0 | 0 | 8 | 321 | 8 |
161 |
8 |
GTM Transceivers (58G (112G))2 | 0 |
0 |
0 |
0 |
0 |
8 (4)2 | 0 | 36 (0) | 0 | 36 (0) |
1. 16 GTYP transceivers are dedicated to the CPM for PCI Express use.
2. 112G supported by combining two GTM transceivers together.
AMD provides a leading software development environment for designing with adaptive SoCs and FPGAs—this includes tools (compilers, simulators, etc.), IP, and solutions.
This environment can reduce development time while allowing developers to achieve high performance per watt. AMD adaptive SoCs & FPGA design tools enable all types of developers from AI scientists, application and algorithm engineers, embedded software developers, and traditional hardware developers to use AMD adaptive computing solutions.
Jump-start your design cycle and achieve fast time-to-market with the proven hardware, software support, tools, design examples, and documentation available for the kit.
Start evaluating Versal Prime series capabilities today with the VMK180 evaluation kit featuring the VM1802 device. Through a host of tools, software, libraries, IP, middleware, and frameworks, adaptive SoCs enable dynamically customizable accelerated computing solutions through industry-standard design flows. The Versal Prime series VMK180 evaluation kit has everything you need to jump-start your designs.
Join the Versal adaptive SoC notification list to receive the latest news and updates.