commit | author | age
|
4783f2
|
1 |
--- |
LL |
2 |
# Default variables that are defined by the playbook that will be deploying these roles |
|
3 |
become_override: False |
|
4 |
ocp_username: opentlc-mgr |
|
5 |
silent: False |
|
6 |
|
|
7 |
# This workshop requires per user 10-12 CPU and 16GB of Memory |
|
8 |
# For a 50 person workshop, you'll need ~40 OCP worker nodes running on with m4.4xlarge instances |
|
9 |
_infra_node_replicas: 40 |
|
10 |
_infra_node_instance_type: m4.4xlarge |
|
11 |
|
|
12 |
user_name: null |
|
13 |
project_name: null |
|
14 |
|
|
15 |
# The first user number to start with when creating projects |
|
16 |
user_count_start: 1 |
|
17 |
# The last user number to start with when creating projects |
852e52
|
18 |
user_count_end: "{{user_count_start|int + num_users|int}}" |
4783f2
|
19 |
|
LL |
20 |
# Rook Ceph Info |
|
21 |
# These variables will be initialized by the role ocp4-workload-rhte-analytics_data_ocp_infra |
|
22 |
# The Rook Ceph RGW service ClusterIP in the rook-ceph namespace |
|
23 |
rgw_service_ip: null |
|
24 |
# The Rook Ceph RGW service port in the rook-ceph namespace |
|
25 |
rgw_service_port: null |
|
26 |
# Full RGW endpoint url that can be used as the S3 endpoint |
|
27 |
rgw_endpoint_url: "http://{{ rgw_service_ip }}:{{ rgw_service_port }}" |
|
28 |
|
|
29 |
# Open Data Hub |
|
30 |
# Deploy the Open Data Hub CustomResource in the user project space, |
|
31 |
# after the project is setup and ODH ClusterServiceVersion has finished |
|
32 |
deploy_odh_cr: false |
|
33 |
# Open Data Hub operator image |
|
34 |
odh_operator_image_repo: "quay.io/opendatahub/opendatahub-operator:v0.4.0" |
|
35 |
# Registry and repistory for AICoE-JupyterHub images |
|
36 |
# EXAMPLE: podman pull {{ jupyterhub_image_registry }}/{{ jupyterhub_image_repository }}:IMAGE:TAG |
|
37 |
jupyterhub_image_registry: 'quay.io' |
|
38 |
jupyterhub_image_repository: 'odh-jupyterhub' |
|
39 |
|
|
40 |
# Custom notebook image source that will be used for the workshop |
|
41 |
workshop_jupyter_notebook_imagestream_image: "quay.io/llasmith/spark-notebook:spark-2.3.2_hadoop-2.8.5" |
|
42 |
# Imagestream name for the custom workshop image |
|
43 |
workshop_jupyter_notebook_imagestream_name: s2i-spark-scipy-notebook |
|
44 |
# Imagestream tag for the custom workshop image |
|
45 |
workshop_jupyter_notebook_imagestream_tag: "3.6" |
|
46 |
|
|
47 |
# Command separated string list each git repo url to preload on the notebook pod when it spawns |
712bd4
|
48 |
workshop_preload_repos: "https://gitlab.com/opendatahub/data-engineering-and-machine-learning-workshop.git" |
4783f2
|
49 |
|
LL |
50 |
# Amount of memory for the JupyterHub server |
|
51 |
jupyterhub_memory: "1Gi" |
|
52 |
# Amount of memory for the spawned Jupyter Notebook pods |
|
53 |
jupyter_notebook_memory: "2Gi" |
|
54 |
|
|
55 |
# Image to use for the Spark Cluster |
|
56 |
spark_node_image: "quay.io/llasmith/openshift-spark:spark-2.3.2_hadoop-2.8.5" |
|
57 |
# Number of Spark master nodes |
|
58 |
spark_master_count: 1 |
|
59 |
# Number of Spark worker nodes |
|
60 |
spark_worker_count: 2 |
|
61 |
# Amount of memory to allocate to each Spark node. This amount will be used for master AND worker nodes |
|
62 |
spark_node_memory: "4Gi" |
|
63 |
# Amount of cpu to allocate to each Spark node. This amount will be used for master AND worker nodes |
|
64 |
spark_node_cpu: 1 |
|
65 |
|
|
66 |
# Path to append to env var PYTHONPATH for pyspark module |
|
67 |
spark_pythonpath: "/opt/app-root/lib/python3.6/site-packages/pyspark/python/:/opt/app-root/lib/python3.6/site-packages/pyspark/python/lib/py4j-0.10.7-src.zip" |
|
68 |
# PySpark submit args to be set as the env var SPARK_SUBMIT_ARGS |
|
69 |
spark_submit_args: "--conf spark.cores.max=1 --conf spark.executor.instances=1 --conf spark.executor.memory=4G --conf spark.executor.cores=1 --conf spark.driver.memory=2G --packages com.amazonaws:aws-java-sdk:1.8.0,org.apache.hadoop:hadoop-aws:2.8.5 pyspark-shell" |