Nimbix Container Flow for Library and Application

This project provides script to build Docker Application for Nimbix Jarvice platform based on docker image.

Usage

1. Clone nimbixlize repository

git clone https://gitenterprise.xilinx.com/FaaSApps/nimbixlize.git

2. Go to nimbixlize repository

cd nimbixlize

3. Modify file AppDef.json.example. Update attribute machines on line 10. Replace __update machine types here__ with machine type id list below.

Machine Type ID	Platform	Mark
n2	N/A	8 core, 64GB RAM (CPU only)
n3	N/A	16 core, 128GB RAM (CPU only)
n4	N/A	16 core, 256GB RAM (CPU only)
n5	N/A	16 core, 512GB RAM (CPU only)
n9	N/A	20 core Intel Skylake, 192GB RAM (CPU only)
nx5u	xilinx_u200_xdma_201820_1	16 core, 128GB RAM, Xilinx Alveo U200 FPGA 2018.2 XDF
nx5u_xdma_201830_1	xilinx_u200_xdma_201830_1	16 core, 128GB RAM, Xilinx Alveo U200 FPGA 2018.3
nx5u_xdma_201830_2	xilinx_u200_xdma_201830_2	16 core, 128GB RAM, Xilinx Alveo U200 FPGA 2019.1
nx5u_xdma_201830_2_2_3	xilinx_u200_xdma_201830_2	16 core, 128GB RAM, Xilinx Alveo U200 FPGA 2019.2
nx6u	xilinx_u250_xdma_201820_1	16 core, 128GB RAM, Xilinx Alveo U250 FPGA 2018.2 XDF
nx6u_xdma_201830_1	xilinx_u250_xdma_201830_1	16 core, 128GB RAM, Xilinx Alveo U250 FPGA 2018.3
nx6u_xdma_201830_2	xilinx_u250_xdma_201830_2	16 core, 128GB RAM, Xilinx Alveo U250 FPGA 2019.1
nx6u_xdma_201830_2_2_3	xilinx_u250_xdma_201830_2	16 core, 128GB RAM, Xilinx Alveo U250 FPGA 2019.2
nx7u_xdma_201920_1	xilinx_u280_xdma_201920_1	16 core, 128GB RAM, Xilinx Alveo U280 FPGA 2019.2

For example:

	"machines": [
        "n2",
        "n3",
        "nx7u_xdma_201920_1"
    ],

4. Build Nimbix image

   4.1 ./utility.sh

   4.2 Enter your origin docker image id:

   4.3 Enter your building Nimbix docker image id :

   4.4 Enter your building Nimbix application name:

   4.5 Enter your buiilding Nimbix application description (optional):

5. Push built docker image

docker push $(IMAGE ID)