Skip to content

Commit

Permalink
[#32562] Incorporate Prism into the Beam Website. (#32563)
Browse files Browse the repository at this point in the history
  • Loading branch information
lostluck committed Sep 30, 2024
1 parent e640b25 commit 9487927
Show file tree
Hide file tree
Showing 11 changed files with 196 additions and 8 deletions.
7 changes: 4 additions & 3 deletions .github/autolabeler.yml
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ io: ["sdks/go/pkg/beam/io/**/*", "sdks/java/io/**/*", "sdks/python/apache_beam/
"mqtt": ["sdks/java/io/mqtt/**/*"]
"parquet": ["sdks/java/io/parquet/**/*"]
"rabbitmq": ["sdks/java/io/rabbitmq/**/*"]
"redis": ["sdks/java/io/redis/**/*"]
"redis": ["sdks/java/io/redis/**/*"]
"solr": ["sdks/java/io/solr/**/*"]
"spanner": ["sdks/go/pkg/beam/io/spannerio/**/*", "sdks/python/apache_beam/io/gcp/spanner.py", "sdks/python/apache_beam/io/gcp/experimental/spannerio.py", "sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/spanner/**/*"]
"bigtable": ["sdks/go/pkg/beam/io/bigtableio/**/*", "sdks/go/pkg/beam/io/xlang/bigtableio/**/*", "sdks/python/apache_beam/io/gcp/bigtableio.py", "sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/bigtable/**/*"]
Expand All @@ -76,9 +76,9 @@ io: ["sdks/go/pkg/beam/io/**/*", "sdks/java/io/**/*", "sdks/python/apache_beam/
"thrift": ["sdks/java/io/thrift/**/*"]
"tika": ["sdks/java/io/tika/**/*"]
"xml": ["sdks/java/io/xml/**/*"]

# Runners
"runners": ["runners/**/*", "sdks/go/pkg/beam/runners/**/*", "sdks/python/runners/**/*", "sdks/typescript/src/apache_beam/runners/**/*"]
"runners": ["runners/**/*", "sdks/go/pkg/beam/runners/**/*", "sdks/python/apache_beam/runners/**/*", "sdks/typescript/src/apache_beam/runners/**/*"]
"core": ["runners/core-construction-java/**/*", "runners/core-java/**/*"]
"dataflow": ["runners/google-cloud-dataflow-java/**/*", "sdks/go/pkg/beam/runners/dataflow/**/*", "sdks/python/runners/dataflow/**/*"]
"direct": ["runners/direct-java/**/*", "sdks/go/pkg/beam/runners/direct/**/*", "sdks/python/runners/direct/**/*"]
Expand All @@ -88,6 +88,7 @@ io: ["sdks/go/pkg/beam/io/**/*", "sdks/java/io/**/*", "sdks/python/apache_beam/
"jet": ["runners/jet/**/*"]
"local": ["runners/local-java/**/*"]
"portability": ["runners/portability/**/*"]
"prism": ["runners/prism/**/*", "sdks/go/pkg/beam/runners/prism/**/*", "sdks/go/cmd/prism/**/*", "sdks/python/apache_beam/runners/portability/prism_runner.py","sdks/python/apache_beam/runners/portability/prism_runner_test.py"]
"samza": ["runners/samza/**/*"]
"spark": ["runners/spark/**/*", "sdks/go/pkg/beam/runners/spark/**/*"]
"twister2": ["runners/twister2/**/*"]
1 change: 1 addition & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -65,6 +65,7 @@ Have ideas for new SDKs or DSLs? See the [sdk-ideas label](https://github.com/ap
Beam supports executing programs on multiple distributed processing backends through PipelineRunners. Currently, the following PipelineRunners are available:

- The `DirectRunner` runs the pipeline on your local machine.
- The `PrismRunner` runs the pipeline on your local machine using Beam Portability.
- The `DataflowRunner` submits the pipeline to the [Google Cloud Dataflow](http://cloud.google.com/dataflow/).
- The `FlinkRunner` runs the pipeline on an Apache Flink cluster. The code has been donated from [dataArtisans/flink-dataflow](https://github.com/dataArtisans/flink-dataflow) and is now part of Beam.
- The `SparkRunner` runs the pipeline on an Apache Spark cluster.
Expand Down
2 changes: 1 addition & 1 deletion website/www/site/content/en/blog/beam-2.59.0.md
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ For more information on changes in 2.59.0, check out the [detailed release notes
## Highlights

* Added support for setting a configureable timeout when loading a model and performing inference in the [RunInference](https://beam.apache.org/documentation/ml/inference-overview/) transform using [with_exception_handling](https://beam.apache.org/releases/pydoc/current/apache_beam.ml.inference.base.html#apache_beam.ml.inference.base.RunInference.with_exception_handling) ([#32137](https://github.com/apache/beam/issues/32137))
* Initial experimental support for using Prism with the Java and Python SDKs
* Initial experimental support for using [Prism](/documentation/runners/prism/) with the Java and Python SDKs
* Prism is presently targeting local testing usage, or other small scale execution.
* For Java, use 'PrismRunner', or 'TestPrismRunner' as an argument to the `--runner` flag.
* For Python, use 'PrismRunner' as an argument to the `--runner` flag.
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -22,9 +22,9 @@ Testing your pipeline is a particularly important step in developing an effectiv

Before running your pipeline on the runner of your choice, unit testing your pipeline code locally is often the best way to identify and fix bugs in your pipeline code. Unit testing your pipeline locally also allows you to use your familiar/favorite local debugging tools.

You can use [DirectRunner](/documentation/runners/direct), a local runner helpful for testing and local development.
You can use [DirectRunner](/documentation/runners/direct), or [PrismRunner](/documentation/runners/prism). Both are local runners helpful for testing and local development.

After you test your pipeline using the `DirectRunner`, you can use the runner of your choice to test on a small scale. For example, use the Flink runner with a local or remote Flink cluster.
After you test your pipeline locally, you can use the runner of your choice to test on a small scale. For example, use the Flink runner with a local or remote Flink cluster.

The Beam SDKs provide a number of ways to unit test your pipeline code, from the lowest to the highest levels. From the lowest to the highest level, these are:

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,7 @@ Using the Direct Runner for testing and development helps ensure that pipelines

Here are some resources with information about how to test your pipelines.
<ul>
<li><a href="/documentation/pipelines/test-your-pipeline/">Test Your Pipeline</a></li>
<!-- Java specific links -->
<li class="language-java"><a href="/blog/2016/10/20/test-stream.html">Testing Unbounded Pipelines in Apache Beam</a> talks about the use of Java classes <a href="https://beam.apache.org/releases/javadoc/{{< param release_latest >}}/index.html?org/apache/beam/sdk/testing/PAssert.html">PAssert</a> and <a href="https://beam.apache.org/releases/javadoc/{{< param release_latest >}}/index.html?org/apache/beam/sdk/testing/TestStream.html">TestStream</a> to test your pipelines.</li>
<li class="language-java">The <a href="/get-started/wordcount-example/#testing-your-pipeline-with-asserts">Apache Beam WordCount Walkthrough</a> contains an example of logging and testing a pipeline with <a href="https://beam.apache.org/releases/javadoc/{{< param release_latest >}}/index.html?org/apache/beam/sdk/testing/PAssert.html">PAssert</a>.</li>
Expand Down
145 changes: 145 additions & 0 deletions website/www/site/content/en/documentation/runners/prism.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,145 @@
---
type: runners
title: "Prism Runner"
aliases: /learn/runners/prism/
---
<!--
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->

# Overview

The Apache Beam Prism Runner can be used to execute Beam pipelines locally using [Beam Portability](/roadmap/portability/).

The Prism runner is suitable for small scale local testing and provides:

* A statically compiled, single binary for simple deployment without additional configuration.
* A web UI when executing in stand alone mode.
* A direct implementation of Beam execution semantics.
* A streaming-first runtime that supports batch processing and data streaming programs.
* Fast, in-memory execution for to simplify SDK, Transform, and Pipeline development.
* Cross Language Transform support.

Written in [Go](https://go.dev), it is the default runner for the [Go SDK](/roadmap/go-sdk/), but can be used in other SDKs as well (see below).

# Capabilities

While Prism already supports a great deal of Beam features, it doesn't yet support everything.
Prism is under active development to close these gaps.

With the exception of timer issues, use of unsupported features should fail the pipeline at job submission time.

In the [2.59.0 release](/blog/beam-2.59.0/), Prism passes most runner validations tests with the exceptions of pipelines using the following features:

OrderedListState, OnWindowExpiry (eg. GroupIntoBatches), CustomWindows, MergingWindowFns, Trigger and WindowingStrategy associated features, Bundle Finalization, Looping Timers, and some Coder related issues such as with Python combiner packing, and Java Schema transforms, and heterogenous flatten coders.
Processing Time timers do not yet have real time support.


See the [Roadmap](/roadmap/prism-runner/) for how to find current progress.
Specific feature support information will soon migrate to the [Runner Capability Matrix](/documentation/runners/capability-matrix/).

# Using the Prism Runner

{{< language-switcher go java py >}}

<span class="language-go">Prism is the default runner for the Go SDK and is used automatically. Set the runner with the flag `--runner=PrismRunner`. </span>
<span class="language-java">Set the runner to `PrismRunner`. </span>
<span class="language-py">Set the runner to `PrismRunner`. </span>

For other SDKs, Prism is included as an asset on [Beam Github Releases](https://github.com/apache/beam/releases/tag/v{{< param release_latest >}}) for download and stand alone use.

Here are some resources with information about how to test your pipelines.
<ul>
<li><a href="/documentation/pipelines/test-your-pipeline/">Test Your Pipeline</a></li>
<li>The <a href="/get-started/wordcount-example/#testing-your-pipeline-with-asserts">Apache Beam WordCount Walkthrough</a> contains an example of logging and testing a pipeline with asserts.
<!-- Java specific links -->
<li class="language-java"><a href="/blog/2016/10/20/test-stream.html">Testing Unbounded Pipelines in Apache Beam</a> talks about the use of Java classes <a href="https://beam.apache.org/releases/javadoc/{{< param release_latest >}}/index.html?org/apache/beam/sdk/testing/PAssert.html">PAssert</a> and <a href="https://beam.apache.org/releases/javadoc/{{< param release_latest >}}/index.html?org/apache/beam/sdk/testing/TestStream.html">TestStream</a> to test your pipelines.</li>
</ul>

### Specify your dependency

<span class="language-java">When using Java, you must specify your dependency on the Direct Runner in your `pom.xml`.</span>
{{< highlight java >}}
<dependency>
<groupId>org.apache.beam</groupId>
<artifactId>beam-runners-prism-java</artifactId>
<version>{{< param release_latest >}}</version>
<scope>runtime</scope>
</dependency>
{{< /highlight >}}

<span class="language-py">This section is not applicable to the Beam SDK for Python. Prism is built in.</span>
<span class="language-go">This section is not applicable to the Beam SDK for Go. Prism is built in.</span>

Except for the Go SDK, Prism is included as an asset on [Beam Github Releases](https://github.com/apache/beam/releases/tag/v{{< param release_latest >}}) for automatic download, startup, and shutdown on SDKs.
The binary is cached locally for subsequent executions.

## Pipeline options for the Prism Runner

Prism aims to have minimal configuration required, and does not currently present user pipeline options.

## Running Prism Standalone

Prism can be executed as a stand alone binary and will present a basic UI for listing jobs, and job status.
This is an optional mode for Prism that is useful for demos or rapid iteration.
It is not a requirement for using Prism in the Java or Python SDKs.

This can be done in two ways, downloading an asset from the github release, or building the binary locally with Go installed.

In either case, Prism serves a JobManagement API endpoint, and a Webpage UI locally.
Jobs can be submitted using `--runner=PortableRunner --endpoint=<endpoint address>` and monitored using the webpage UI.

Example output from the Prism binary:

```
2024/09/30 09:56:42 INFO Serving JobManagement endpoint=localhost:8073
2024/09/30 09:56:42 INFO Serving WebUI endpoint=http://localhost:8074
```

The binary has the following optional flags:

* `--job_port` sets the port for the Job management server (defaults to 8073)
* `--web_port` sets the port for the web ui (defaults to 8074)
* `--serve_http` enables or disables the web ui (defaults to true)
* `---idle_shutdown_timeout` sets a duration that Prism will wait for a new job before automatically shutting itself down. Uses duration format like `10s`, `5m`,`2h`. Defaults to not shutting down.

### Download a release asset

This approach doesn't require other dependencies or runtimes installed.
This is recommended if you want to deploy Prism on some other machine.

Navigate to the latest [Beam Release Github page](https://github.com/apache/beam/releases/tag/v{{< param release_latest >}}), scroll to the bottom, and download the correct asset for where you want to run Prism.

For example, if you want to execute Prism on a newer MacBook, you'd download the `darwin-arm64` asset. For executing on many cloud machines, you'd download the `linux-amd64` asset.

This requires downloading the right asset for the machine Prism will run on, such as your development machine.

Simply unzip, and execute.

### Build from the release with Go.

This approach requires a [recent version of Go installed](https://go.dev/dl/).
This is recommended if you only want to run Prism on your local machine.

You can insall Prism with `go install`:

```sh
go install github.com/apache/beam/sdks/v2/go/cmd/prism@latest
prism
```

Or simply build and execute the binary immeadiately using `go run`:

```sh
go run github.com/apache/beam/sdks/v2/go/cmd/prism@latest
```
4 changes: 2 additions & 2 deletions website/www/site/content/en/roadmap/_index.md
Original file line number Diff line number Diff line change
Expand Up @@ -32,8 +32,8 @@ Below are some highlights for the project as a whole.
Portability is the primary Beam vision: running pipelines authored with _any SDK_
on _any runner_. This is a cross-cutting effort across Java, Python, and Go,
and every Beam runner. Portability is currently supported on the
[Flink](/documentation/runners/flink/)
and [Spark](/documentation/runners/spark/) runners.
[Flink](/documentation/runners/flink/), [Spark](/documentation/runners/spark/)
and [Prism](/documentation/runners/prism/) runners.

See the details on the [Portability Roadmap](/roadmap/portability/)

Expand Down
35 changes: 35 additions & 0 deletions website/www/site/content/en/roadmap/prism-runner.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,35 @@
---
title: "Prism Runner Roadmap"
---
<!--
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->

# Apache Beam Prism Runner Roadmap

The goal for the Prism runner is to provide a good default onboarding experience for Apache Beam.

* Prism should be able to execute any Beam pipeline that can execute on a local machine.
* Prism should be fast to start and execute pipelines.
* Prism should be able to assist with the local testing and debugging of pipelines.
* Prism may develop into a robust, production ready runner for pipelines that can execute locally.

The detailed roadmap lives in an [umbrella tracking issue in Github](https://github.com/apache/beam/issues/29650).

Here are available resources:

- [Runner documentation](/documentation/runners/prism)
- Issues: [prism](https://github.com/apache/beam/issues?q=is%3Aopen+is%3Aissue+label%3Aprism)
- CLI [Code](https://github.com/apache/beam/tree/master/sdks/go/cmd/prism), CLI Binaries are available as assets on [Github Releases](https://github.com/apache/beam/releases/tag/v{{< param release_latest >}}).
- Core [Code](https://github.com/apache/beam/tree/master/sdks/go/pkg/beam/runners/prism)
- [Prism Internals Deep Dive](https://github.com/apache/beam/blob/master/sdks/go/pkg/beam/runners/prism/internal/README.md)
3 changes: 3 additions & 0 deletions website/www/site/data/en/documentation_runners.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,9 @@
- name: { text: "DirectRunner:", link: /documentation/runners/direct/ }
icon: icons/documentation/runners/beam-icon.svg
description: Runs locally on your machine – great for developing, testing, and debugging.
- name: { text: "PrismRunner:", link: /documentation/runners/prism/ }
icon: icons/documentation/runners/beam-icon.svg
description: Runs locally on your machine – great for developing, testing, and debugging.
- name: { text: "FlinkRunner:", link: /documentation/runners/flink/ }
icon: icons/documentation/runners/flink-icon.svg
description: Runs on <a target="_blank" href="https://flink.apache.org">Apache Flink</a>.
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,7 @@

<ul class="section-nav-list">
<li><a href="/roadmap/dataflow-runner/">Google Cloud Dataflow Runner</a></li>
<li><a href="/roadmap/prism-runner/">Prism Runner</a></li>
<li><a href="/roadmap/flink-runner/">Flink Runner</a></li>
<li><a href="/roadmap/nemo-runner/">Nemo Runner</a></li>
<li><a href="/roadmap/spark-runner/">Spark Runner</a></li>
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@
<li><span class="section-nav-list-main-title">Runners</span></li>
<li><a href="/documentation/runners/capability-matrix/">Capability Matrix</a></li>
<li><a href="/documentation/runners/direct/">Direct Runner</a></li>
<li><a href="/documentation/runners/prism/">Prism Runner</a></li>
<li><a href="/documentation/runners/flink/">Apache Flink</a></li>
<li><a href="/documentation/runners/nemo/">Apache Nemo</a></li>
<li><a href="/documentation/runners/samza/">Apache Samza</a></li>
Expand Down

0 comments on commit 9487927

Please sign in to comment.