MyAdvantech Registration

MyAdvantech is a personalized portal for Advantech customers. By becoming an Advantech member, you can receive latest product news, webinar invitations and special eStore offers.

Sign up today to get 24/7 quick access to your account information.

VEGA-550
Reconfigurable Video Content Intelligence Accelerator

  • Four high flexible Xilinx Zynq UltraScale+ ZU7EV MPSoC
  • 8GB+8GB DDR4 w/ ECC for each FPGA device
  • PCIe Gen 3 x16 host interface
  • GPU card size form factor
  • Up to 150W slot power consumption
As more and more video analytics solutions has been used in different application. Advantech VEGA-500 series product provides an ideal platform with video codec and programmable logics integrated for video intelligence applications. VEGA-550 can decode/encode up to 32-ch 1080p30 video streams simultaneously and pass the decoded video to programmable logic for further video intelligence applications.

Product Specifications Download Spec

FPGA
  • Zynq UltraScale+ XCZU7EV MPSoC
Memory
  • DDR4
Storage
  • eMMC
I/O
  • QSPI
  • Ethernet port (RJ45)
  • USB3.0
  • USB console (UART to USB)
  • JTAG
  • PCIe
Integrated IP
  • Video Codec Unit (VCU)
Power
  • Power Consumption
  • Ext. power connector
Physical Dimensions
  • FH/FL double slots

FPGA

  • Zynq UltraScale+ XCZU7EV MPSoC

    VEGA-550S-I2A0 4 FPGA devices
    VEGA-550L-I2A0 4 FPGA devices

Memory

  • DDR4

    VEGA-550S-I2A0 8GB DDR4-2400 64-bit + ECC to processing system(PS) per FPGA
    8GB DDR4-2666 64-bit + ECC to programmable logic(PL) per FPGA
    VEGA-550L-I2A0 8GB DDR4-2400 64-bit + ECC to processing system(PS) per FPGA
    8GB DDR4-2666 64-bit + ECC to programmable logic(PL) per FPGA

Storage

  • eMMC

    VEGA-550S-I2A0 8GB x 1 per FPGA
    VEGA-550L-I2A0 8GB x 1 per FPGA

I/O

  • QSPI

    VEGA-550S-I2A0 32MB x 1 per FPGA
    VEGA-550L-I2A0 32MB x 1 per FPGA
  • Ethernet port (RJ45)

    VEGA-550S-I2A0 1 to FPGA0
    VEGA-550L-I2A0 1 to FPGA0
  • USB3.0

    VEGA-550S-I2A0 1 to each FPGA
    VEGA-550L-I2A0 1 to each FPGA
  • USB console (UART to USB)

    VEGA-550S-I2A0 1 per FPGA
    VEGA-550L-I2A0 1 per FPGA
  • JTAG

    VEGA-550S-I2A0 1 per FPGA
    VEGA-550L-I2A0 1 per FPGA
  • PCIe

    VEGA-550S-I2A0 Gen 3 x 16 to host
    VEGA-550S (More PCIe to PS side):
    PCIe Gen 3 x 4 to PL side in FPGA 0 & 1
    PCIe Gen 3 x 2 to PL side in FPGA 2 & 3 PCIe Gen 2 x 4 to PS side per FPGA
    VEGA-550L-I2A0 Gen 3 x 16 to host
    VEGA-550L (More PCIe to PL side):
    PCIe Gen 3 x 4 to PL side per FPGA
    PCIe Gen 2 x 4 to PS side in FPGA 0 & 1

Integrated IP

  • Video Codec Unit (VCU)

    VEGA-550S-I2A0 4:2:2 10-bit 4Kp60 per FPGA
    VEGA-550L-I2A0 4:2:2 10-bit 4Kp60 per FPGA

Power

  • Power Consumption

    VEGA-550S-I2A0 <150W
    VEGA-550L-I2A0 <150W
  • Ext. power connector

    VEGA-550S-I2A0 2x3 pin
    VEGA-550L-I2A0 2x3 pin

Physical Dimensions

  • FH/FL double slots

    VEGA-550S-I2A0 111 x 267 mm
    VEGA-550L-I2A0 111 x 267 mm

Need more info? Talk to our Online Rep!

Featured Highlights

  • FPGA Platforms for Fastest Deep Learning Inference

    FPGA Platforms for Fastest Deep Learning Inference

    Advantech‘s VEGA Series of Xilinx FPGA-based PCI Express cards have been designed to accelerate AI deep learning workloads, offering a range of integration options to suit customer needs. In particular, the range focuses on the needs of video-based, real-time AI inference applications at the edge where fastest video processing is critical.