sdeoras

Updated 11/28/2018

Here's my experience of installing the NVIDIA CUDA kit 9.0 on a fresh install of Ubuntu Desktop 16.04.4 LTS. Scroll down to the bottom if you wish to only install NVIDIA drivers and run tensorflow via docker container.

Table of Contents

• Install NVIDIA Graphics Driver via apt-get
• Install CUDA
• Install cuDNN

Table of contents generated with markdown-toc

sdeoras

Follow instructions here: https://www.tensorflow.org/install/gpu#ubuntu_1804_cuda_10
Download and install Anaconda with Python 3.7: https://www.anaconda.com/distribution/#linux
pip install --upgrade tensorflow-gpu
python -c "import tensorflow as tf; tf.enable_eager_execution(); print(tf.reduce_sum(tf.random_normal([1000, 1000])))"

sdeoras

Steps to build TensorFlow v1.12.0 C-library on Jetson Nano for GPU

Jetson Nano configuration

• ARM 64 (aarch64)
• gcc 7.3
• cuda 10
• cudnn 7

TensorFlow configuration

• v1.12.0

sdeoras

podman --runtime=runsc run --rm -it docker.io/library/ubuntu:latest bash
I0604 20:47:56.621539   21217 x:0] ***************************
I0604 20:47:56.621611   21217 x:0] Args: [/usr/bin/runsc start fb6738612f208a2786470ab33803763b976290531168c3716ea72b30ae74f310]
I0604 20:47:56.621673   21217 x:0] Version 90a116890fce
I0604 20:47:56.621686   21217 x:0] PID: 21217
I0604 20:47:56.621698   21217 x:0] UID: 0, GID: 0
I0604 20:47:56.621706   21217 x:0] Configuration:
I0604 20:47:56.621712   21217 x:0] 		RootDir: /run/user/1000/runsc
I0604 20:47:56.621720   21217 x:0] 		Platform: ptrace
I0604 20:47:56.621734   21217 x:0] 		FileAccess: exclusive, overlay: false

sdeoras

create sandboxed cluster

please set PROJECT env. var to your Google Cloud project name.

gcloud beta container \
    --project "${PROJECT}" \
    clusters create "sandboxed" \
    --zone "us-west1-a" \
    --no-enable-basic-auth \
    --cluster-version "1.13.7-gke.24" \

sdeoras

Steps required to setup auto-unseal of vault using GCP KMS on k8s are as follows:

• Create a keyring and key
• Generate a service account credentials with encrypter-decryper role
• Download https://github.com/hashicorp/vault-helm
• Checkout a particular version (say tags/v0.3.0)
• Edit values.yaml and update it as shown below.
• Install helm/tiller and install vault component

diff for values.yaml

sdeoras

path "auth/userpass/users/+" {
  capabilities = ["create", "read", "delete", "update", "list"]
  allowed_parameters = {
    "token-policies" = ["default", "policies/*"]
    "policies" = ["default", "policies/*"]
    "*" = []
  }
}

path "auth/userpass/users/+/policies" {

sdeoras

apiVersion: v1
kind: Namespace
metadata:
  name: julia-system
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: julia
  labels:

sdeoras

apiVersion: dapr.io/v1alpha1
kind: Component
metadata:
  annotations:
    meta.helm.sh/release-name: sensor-config
    meta.helm.sh/release-namespace: sensor-system
  creationTimestamp: "2021-11-30T18:00:19Z"
  generation: 1
  labels:
    app.kubernetes.io/managed-by: Helm

sdeoras

apiVersion: influxdb2.deoras-labs.io/v1beta1
kind: Config
metadata:
  name: default
  namespace: sensor-system
spec:
  addr: http://influxdb2.influxdb2-system.svc.cluster.local
  orgName: sensor-system
  tokenSecretName: admin-token
  tokenSecretNamespace: sensor-system