Add recurring run demo (#917)

* Add recurring run demo

Author: droctothorpe

demos/README.md

@@ -11,3 +11,7 @@ that's suitable for presentation to public audiences.
 
 * [Simple pipeline](simple_pipeline/): highlights the use of pipelines and
   hyperparameter tuning on a GKE cluster with node autoprovisioning.
+
+* [Recurring Run](recurring_run/): A  simple demo that illustrates how to use
+  the Kubeflow Pipelines SDK to provision [recurring
+  runs](https://www.kubeflow.org/docs/components/pipelines/concepts/run/).

demos/recurring/README.md

@@ -0,0 +1,27 @@
+# Kubeflow demo - Recurring runs with the KFP SDK
+
+## 1. Setup your environment
+
+This demo assumes that you have a functioning Kubeflow Pipelines deployment. If
+not, follow the instructions
+[here](https://www.kubeflow.org/docs/components/pipelines/installation/) and
+[here](https://www.kubeflow.org/docs/components/pipelines/sdk/install-sdk/).
+
+This demo has been verified to work with:
+- KFP version `1.7.1`
+- KFP SDK version `1.8.11`
+
+Activate the conda environment you created following the above steps. 
+
+Create a Jupyter kernel for your conda environment.
+
+```bash
+ipython kernel install --name "kfp" --user
+```
+
+## 2. Run the KFP SDK script
+
+Step through the provided [notebook](recurring.ipynb) to create a recurring run
+using the KFP SDK. Make sure to select the `kfp` kernel that you created
+earlier. 
+

demos/recurring/component.yaml

@@ -0,0 +1,19 @@
+name: Download
+inputs:
+- {name: Url, type: URI}
+metadata:
+  annotations:
+    author: Alexander Perlman <[email protected]>
+implementation:
+  container:
+    image: alpine/curl 
+    command:
+    - sh
+    - -exc
+    - |
+      url="$0"
+      path='/tmp/script'
+      curl "$url" -o "$path"
+      chmod 700 "$path"
+      /bin/sh "$path"
+    - inputValue: Url

demos/recurring/download.yaml

@@ -0,0 +1,56 @@
+apiVersion: argoproj.io/v1alpha1
+kind: Workflow
+metadata:
+  generateName: pipeline-
+  annotations: {pipelines.kubeflow.org/kfp_sdk_version: 1.8.11, pipelines.kubeflow.org/pipeline_compilation_time: '2022-01-31T21:51:10.599476',
+    pipelines.kubeflow.org/pipeline_spec: '{"inputs": [{"name": "url"}], "name": "Pipeline"}'}
+  labels: {pipelines.kubeflow.org/kfp_sdk_version: 1.8.11}
+spec:
+  entrypoint: pipeline
+  templates:
+  - name: download
+    container:
+      args: []
+      command:
+      - sh
+      - -exc
+      - |
+        url="$0"
+        path='/tmp/script'
+        curl "$url" -o "$path"
+        chmod 700 "$path"
+        /bin/sh "$path"
+      - '{{inputs.parameters.url}}'
+      image: alpine/curl
+    inputs:
+      parameters:
+      - {name: url}
+    metadata:
+      annotations: {author: Alexander Perlman <[email protected]>, pipelines.kubeflow.org/component_spec: '{"implementation":
+          {"container": {"command": ["sh", "-exc", "url=\"$0\"\npath=''/tmp/script''\ncurl
+          \"$url\" -o \"$path\"\nchmod 700 \"$path\"\n/bin/sh \"$path\"\n", {"inputValue":
+          "Url"}], "image": "alpine/curl"}}, "inputs": [{"name": "Url", "type": "URI"}],
+          "metadata": {"annotations": {"author": "Alexander Perlman <[email protected]>"}},
+          "name": "Download"}', pipelines.kubeflow.org/component_ref: '{"digest":
+          "1bb47e384d056817b16202398d1e5fc8ce02daf1e40f69e3103218402c05437b", "url":
+          "https://raw.githubusercontent.com/droctothorpe/examples/master/demos/recurring/component.yaml"}',
+        pipelines.kubeflow.org/arguments.parameters: '{"Url": "{{inputs.parameters.url}}"}'}
+      labels:
+        pipelines.kubeflow.org/kfp_sdk_version: 1.8.11
+        pipelines.kubeflow.org/pipeline-sdk-type: kfp
+        pipelines.kubeflow.org/enable_caching: "true"
+  - name: pipeline
+    inputs:
+      parameters:
+      - {name: url}
+    dag:
+      tasks:
+      - name: download
+        template: download
+        arguments:
+          parameters:
+          - {name: url, value: '{{inputs.parameters.url}}'}
+  arguments:
+    parameters:
+    - {name: url}
+  serviceAccountName: pipeline-runner