This document provides the step-by-step procedure to enable Istio integration for NVIDIA Dynamo on OpenShift. Tested with Dynamo 1.4.2 on OpenShift 4.21.
NVIDIA Dynamo is an inference serving platform that provides two Kubernetes CRD APIs for deploying models:
| # place this in .bashrc | |
| function idea() { | |
| ideapath="/opt/idea-IU-223.8617.56/bin/idea.sh" | |
| logfile="$HOME/idea.log" # use /dev/null for no logs | |
| if [ -z "$1" ]; then | |
| ( "$ideapath" > "$logfile" 2>&1 & ) 2> /dev/null | |
| else | |
| ( "$ideapath" "$1" > "$logfile" 2>&1 & ) 2> /dev/null | |
| fi |
| #!/bin/bash | |
| set -e | |
| set -u | |
| set -o pipefail | |
| ######################################## DESCRIPTION ############################################################# | |
| ## Use this script to collect multiple resources into YAML files. ## | |
| ## ## | |
| ## The mandatory '--exp' option takes a grep expression to match against all 'kind.group'. ## |
| ISC License | |
| Copyright (c), Tomer Figenblat | |
| Permission to use, copy, modify, and/or distribute this software for any | |
| purpose with or without fee is hereby granted, provided that the above | |
| copyright notice and this permission notice appear in all copies. | |
| THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES | |
| WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF |
The following are a recording of the steps performed for verifying the feasibility of [Serverless Framework][serverless-framework-site] in a [Knative][knative-docs]-based [OpenShift][openshift-site] environment.
| Serverless Framework 3.21.0 |
| const { graphql } = require('@octokit/graphql'); | |
| // UPDATE the OWNER and the REPOSITORY constants | |
| const OWNER = "Tomerfi"; | |
| const REPOSITORY = "aioswitcher"; | |
| // max 100 for all, for more - use pagination leveraging edges, pageInfo, and cursor. | |
| const TOTAL_FORKS_TO_RETRIEVE = 100; | |
| const TOTAL_BRANCHES_PER_FORK = 100; | |
| const TOTAL_COMMITS_PER_BRANCH = 10; // be careful with this, you might exceed the maximum limit of 500,000 possible nodes. |
| # https://avatars.dicebear.com/ | |
| # https://avatars.dicebear.com/styles/avataaars | |
| dicebear create avataaars --format "png" --style "transparent" \ | |
| --top "shortFlat" --hairColor "blonde" --topChance 100 \ | |
| --accessoriesChance 0 \ | |
| --facialHair "beardMedium" --facialHairColor "blonde" --facialHairChance 100 \ | |
| --clothes "hoodie" --clothesColor "heather" \ | |
| --eyes "squint" --eyebrow "defaultNatural" \ | |
| --mouth "serious" --skin "pale" |
| #!/bin/bash | |
| ##################################################################################################### | |
| # This script is meant to be used as a global smudge-clean filter for removing sensitive data # | |
| # from your commits. # | |
| # # | |
| # 1. Place this script in an acceisble path, i.e. ~/scripts/git-smudge-clean-filter.sh. # | |
| # # | |
| # 2. Populate the 'mapArr' using what you need hidden as the key, and the replacment as the value. # | |
| # DO NOT use same values for multiple keys (this will work only in one direction). # |
| # blog-post for this is here: https://dev.to/tomerfi/my-ultimate-development-dockerfile-4hg1 | |
| # Base image is python's image | |
| FROM python:3.8.5 | |
| # Set shell with pipeline fail return | |
| SHELL ["/bin/bash", "-o", "pipefail", "-c"] | |
| # Set environment variables | |
| ENV SHELL=/bin/bash \ | |
| DOTNET_CLI_TELEMETRY_OPTOUT=true \ |
