Import system images

System-level container images are used by all users. You can obtain images from Lenovo salesperson or import images into LiCO as system-level container images. This section describes how to create and import system-level container images.

Image bootstrap files

LiCO is released with image bootstrap files for commonly-used AI frameworks. The image bootstrap files for commonly-used AI frameworks in the compressed package you obtained(https://hpc.lenovo.com/lico/downloads/6.0/images/k8s/image_bootstrap.zip). Users can use these files to create images.

The table below lists the image bootstrap files.

File nameImagesFrameworkCPU/GPUComments
caffe-1.0-cpucaffe:caffe-1.0-cpuCaffeCPU 
caffe-1.0-gpu-cuda92caffe:caffe-1.0-gpu-cuda92CaffeCUDA 9.2Supports P100 and V100 Caffe does not support CUDA 9.0 officially
chainer-6.7.0-gpu-cuda101chainer:chainer-6.7.0-gpu-cuda101ChainerCUDA 10.1Supports P100, V100, RTX5000, RTX8000 and T4
intel-caffe-1.1.3-cpuintel-caffe:intel-caffe-1.1.3-cpuIntel-caffeCPU 
intel-pythonintel-python:intel-pythonOtherCPU 
jupyter-py36-cpujupyter:jupyter-py36-cpuJupyterCPU 
jupyter-py36-gpujupyter:jupyter-py36-gpuJupyterCUDA 10.0Supports P100, V100, RTX5000, RTX8000 and T4
jupyter-py37-cpujupyter:jupyter-py37-cpuJupyterCPU 
jupyter-py37-gpujupyter:jupyter-py37-gpuJupyterCUDA 10.0Supports P100, V100, RTX5000, RTX8000 and T4
letrain-1.3-cpuletrain:letrain-1.3-cpuLeTrainCPU 
letrain-1.3-gpu-cuda100letrain:letrain-1.3-gpu-cuda100LeTrainCPUSupports P100, V100, RTX5000, RTX8000 and T4
lico-ai-scriptslico-k8s-client:latestOtherCPUIndispensable
lico-file-managerlico-file-manager:latestOtherCPUIndispensable
mxnet-1.5.0-cpu-mklmxnet:mxnet-1.5.0-cpu-mklMxnetCPU 
mxnet-1.5.0-gpu-mkl-cuda100mxnet:mxnet-1.5.0-gpu-mkl-cuda100MxnetCUDA 10.0Supports P100, V100, RTX5000, RTX8000 and T4
neon-2.6-cpuneon:neon-2.6-cpuNeonCPU 
pytorch-1.1.0-gpu-cuda100pytorch:pytorch-1.1.0-gpu-cuda100PyTorchCUDA 10.0Supports P100, V100, RTX5000, RTX8000 and T4
scikit-single-cpuscikit:scikit-single-cpuScikitCPU 
tensorflow-1.15.2-cputensorflow:tensorflow-1.15.2-cpuTensorFlowCPU 
tensorflow-1.15.2-gpu-cuda100tensorflow:tensorflow-1.15.2-gpu-cuda100TensorFlowCUDA 10.0Supports P100, V100, RTX5000, RTX8000 and T4
tensorflow-1.15.2-gpu-cuda100-hbasetensorflow:tensorflow-1.15.2-gpu-cuda100-hbaseTensorFlowCUDA 10.0Supports HBase Supports P100, V100, RTX5000, RTX8000 and T4
tensorflow-1.15.2-gpu-cuda100-kerastensorflow:tensorflow-1.15.2-gpu-cuda100-kerasTensorFlowCUDA 10.0Supports Keras(2.2.4) Supports P100, V100, RTX5000, RTX8000 and T4
tensorflow-1.15.2-gpu-cuda100-mongodbtensorflow:tensorflow-1.15.2-gpu-cuda100-mongodbTensorFlowCUDA 10.0Supports MongoDB Supports P100, V100, RTX5000, RTX8000 and T4
tensorflow-1.15.2-mkltensorflow:tensorflow-1.15.2-mklTensorFlowCPU 
tensorflow-2.1.0-cputensorflow:tensorflow-2.1.0-cpuTensorFlowCPU 
tensorflow-2.1.0-gpu-cuda100tensorflow:tensorflow-2.1.0-gpu-cuda100TensorFlowCUDA 10.0Supports P100, V100, RTX5000, RTX8000 and T4

Create images

Step 1. Check to ensure that the lico-auth-internal.key file is under /etc/lico on the LiCO node.

Step 2. Prepare a build node with a minimum storage of 100 GB.

Notes:

Step 3. To the build node, copy the /etc/lico/lico-auth-internal.key file from the LiCO node. For example, copy the lico-auth-internal.key file to the new directory /opt/images. If the new directory cannot be found, create it manually.

step 4. To the build node, upload the compressed image bootstrap file you obtained which named image_bootstrap.zip(https://hpc.lenovo.com/lico/downloads/6.0/images/k8s/image_bootstrap.zip). For example, upload the compressed package to the new directory /opt/images.

Step 5. To the build node, run the following commands to configure the image:

Step 6. To the build node, do one of the following to create image.

Step 7. Push the created docker image to one existing docker repository, the repository can be one docker registry (https://docs.docker.com/registry/) or one docker harbor (https://goharbor.io/) or docker hub (https://hub.docker.com/), just make sure the k8s nodes can access the docker repository. For example:

Import images into LiCO as system-level images

Run the following commands on the LiCO node to import images from docker repository:

Attention: Modify 10.240.212.106:5000 to the actual url of your docker repository.