This topic describes the features of field programmable gate array (FPGA)-accelerated compute optimized instance families and lists the instance types of each family.

f3, FPGA-accelerated compute optimized instance family

Features
  • Uses Xilinx 16nm Virtex UltraScale+ VU9P FPGAs.
  • Compute:
    • Offers a CPU-to-memory ratio of 1:4.
    • Uses 2.5 GHz Intel® Xeon® Platinum 8163 (Skylake) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports only standard SSDs and ultra disks.
  • Network:

    Provides high network performance based on large computing capacity.

  • Suits the following scenarios:
    • Deep learning and inference
    • Genomics research
    • Database acceleration
    • Image transcoding such as conversion of JPEG images to WebP images
    • Real-time video processing such as H.265 video compression
Instance types
Instance type vCPUs Memory (GiB) FPGAs Bandwidth (bidirectional), Gbit/s Packet forwarding rate (bidirectional), Kpps NIC queues ENIs (including one primary ENI) Private IP addresses per ENI
ecs.f3-c4f1.xlarge 4 16.0 1 × Xilinx VU9P 1.5 300 2 3 10
ecs.f3-c8f1.2xlarge 8 32.0 1 × Xilinx VU9P 2.5 500 4 4 10
ecs.f3-c16f1.4xlarge 16 64.0 1 × Xilinx VU9P 5.0 1,000 4 8 20
ecs.f3-c16f1.8xlarge 32 128.0 2 × Xilinx VU9P 10.0 2,000 8 8 20
ecs.f3-c16f1.16xlarge 64 256.0 4 × Xilinx VU9P 20.0 2,500 16 8 20
ecs.f3-c22f1.22xlarge 88 336.0 4 × Xilinx VU9P 30.0 4,500 16 8 20
Note