Skip to content

Latest commit

 

History

History

stateful-nodecardinality-extraction

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 
 
 
 
 

A Stateful Processing Compiler to Process Input and Feedback Data

This Data Processing Library Scala example shows how to build a stateful processing compiler that extracts the cardinality of nodes in the "Road Topology & Geometry" layer of HERE Map Content catalog and how to use the output of the previous compilation as feedback input to count the number of times the node cardinalities changed for each partition.

The node cardinality refers to the number of segment references for every topology node. In this example we use the Road Topology and Geometry Layer that consists of Topology Geometry Partitions where each Partition contains a set of Nodes .

In this example, the DirectMToNCompiler functional pattern is used, since it allows to implement transformation partitions.

The compiler runs twice with different HERE Map Content catalog versions and does the following:

  • takes HERE Map Content input data as well as data from a previous compilation as feedback
  • finds the cardinality for each node in a partition
  • if the input data has changed since the last compilation, the compiler updates the cardinality and increments the variable that counts the number of times the compiler has run

For more information about the DirectMToNCompiler functional pattern, see the Data Processing Library Developer Guide.

Get Your Credentials

To run this example, you need two sets of credentials:

  • Platform credentials: To get access to the platform data and resources, including HERE Map Content data for your pipeline input.
  • Repository credentials: To download HERE Data SDK for Java & Scala libraries and Maven archetypes to your environment.

For more details on how to set up your credentials, see the Identity & Access Management Developer Guide.

For more details on how to verify that your platform credentials are configured correctly, see the Verify Your Credentials tutorial.

Build and Run the Compiler

In the commands that follow, replace the variable placeholders with the following values:

  • $CATALOG_ID is your output catalog's ID.
  • $CATALOG_HRN is your output catalog's HRN (returned by the olp catalog create command).
  • $PROJECT_HRN is your project's HRN (returned by the olp project create command).
  • $CATALOG_RIB is the HRN of the public HERE Map Content catalog in your pipeline configuration (HERE environment.

Note: We recommend that you set values to variables, so that you can easily copy and execute the following commands.

Run the Compiler Locally

Create a Local Stateful Processing Compiler Catalog

The catalog you create is used to store the cardinality of nodes and the number of times the compiler has run.

To run this compiler locally, use a local output catalog as described below. For more information about local catalogs, see the SDK tutorial about local development and testing and the OLP CLI documentation.

  1. Use the olp local catalog create command to create the local catalog.
olp local catalog create stateful-compiler-scala stateful-compiler-scala --summary "Stateful compiler example catalog" \
            --description "Stateful compiler example catalog"

The local catalog will have the HRN hrn:local:data:::stateful-compiler-scala.

  1. Use the olp local catalog layer add command to add two versioned layers to your catalog:
Layer ID Partitioning Zoom Level Layer Type Content Type Schema Content Encoding
nodecardinality-count HEREtile 12 Versioned application/json None uncompressed
state Generic N.A. Versioned application/octet-stream None uncompressed
olp local catalog layer add hrn:local:data:::stateful-compiler-scala nodecardinality-count nodecardinality-count --versioned --summary "nodecardinality count" \
            --description "nodecardinality count" --partitioning heretile:12 \
            --content-type application/json
olp local catalog layer add hrn:local:data:::stateful-compiler-scala state state --versioned --summary "state" --description "state" \
            --partitioning Generic --content-type application/octet-stream

Build the Compiler

To build the compiler, run sbt package in the stateful-nodecardinality-extraction directory.

sbt package

Run the Compiler from the Command Line

To run the compiler locally, you will need to run the entry point to the compiler:

  • com.here.platform.data.processing.example.scala.feedback.Main

As arguments, you must provide the -Dspark.master parameter with the address of the Spark server master to connect to, and any configuration parameters you want to override. Alternatively, you can add those parameters to the application.conf file.

Additionally, you also need to specify the -Dpipeline-config.file and -Dpipeline-job.file parameters to define the location of a configuration file that contains the catalogs as well as job-specific versions of the catalogs, to read and write to.

The example project provides two template job configurations, config/here/pipeline-job-first.conf and config/here/pipeline-job-second.conf for the first and second run of the pipeline, respectively.

pipeline-job-first.conf specifies in the line version = 1 that the version 1 of the input catalog should be processed in the first run. You can change this version to any number between 0 and the most recent version of the HERE Map Content catalog. You can find the most recent version by opening the HERE platform portal and navigating to the HERE Map Content catalog, and viewing the current catalog's version in the Catalog info section.

pipeline-job-second.conf specifies in the line version = 2 that version 2 of the input catalog should be processed in the second run. You can change this version to any number that is less than or equal to the most recent version of the HERE Map Content catalog and greater than the version specified in config/here/pipeline-job-first.conf.

For local runs, a bounding box filter is provided in the config/here/local-application.conf to limit the number of partitions to be processed. This speeds up the compilation process. In this example, we use a bounding box around the cities of Berlin. You can edit the bounding box coordinates to compile a different partition of HERE Map Content. Make sure you update the layer coverage to reflect the different geographical region. In order to use this configuration file, you need to use the -Dconfig.file parameter.

Set the environment variable $PATH_TO_CONFIG_FOLDER to ./config/here.

Finally, execute the following command in the stateful-nodecardinality-extraction directory to run The Stateful Processing Compiler.

For the HERE platform environment:

sbt run \
-Dpipeline-config.file=./config/here/local-pipeline-config.conf \
-Dpipeline-job.file=./config/here/pipeline-job-first.conf \
-Dconfig.file=./config/here/local-application.conf \
-Dspark.master=local[*]

To observe the behavior of the Stateful Processing Compiler, you have to run the compiler again using pipeline-job-second.conf as job configuration:

For the HERE platform environment:

sbt run \
-Dpipeline-config.file=./config/here/local-pipeline-config.conf \
-Dpipeline-job.file=./config/here/pipeline-job-second.conf \
-Dconfig.file=./config/here/local-application.conf \
-Dspark.master=local[*]

After the second run, in the HERE platform environment, you can inspect the local catalog with the OLP CLI:

olp local catalog inspect hrn:local:data:::stateful-compiler-scala

You should see the nodes with their cardinality, for example here:cm:node:1002845384: 3, where here:cm:node:1002845384 is the node, and 3 - cardinality. The field updatesCount will be 1 for those partitions that did not change during the second run, otherwise it will be 2.

Local Data Inspector

Run this Compiler as a HERE Platform Pipeline

Configure a Project

To follow this example, you will need a project. A project is a collection of platform resources (catalogs, pipelines, and schemas) with controlled access. You can create a project through the HERE platform portal.

Alternatively, use the OLP CLI olp project create command to create the project:

olp project create $PROJECT_ID $PROJECT_NAME

The command returns the HERE Resource Name (HRN) of your new project. Note down this HRN as you will need it later in this tutorial.

Note

You do not have to provide a --scope parameter if your app has a default scope. For details on how to set a default project scope for an app, see the Specify a default Project for Apps chapter of the Identity & Access Management Developer Guide.

For more information on how to work with projects, see the Organize your work in projects tutorial.

Create a Stateful Processing Compiler Catalog

As mentioned above, we will use the public HERE Map Content input catalog, however, we need to create our own output catalog to store the cardinality of nodes and the number of times the compiler has run.

  1. Use the olp catalog create command to create the catalog. Make sure to note down the HRN returned by the following command for later use:
olp catalog create $CATALOG_ID $CATALOG_ID --summary "Stateful compiler example catalog" \
            --description "Stateful compiler example catalog" \
            --scope $PROJECT_HRN
  1. Use the olp catalog layer add command to add two versioned layers to your catalog:
olp catalog layer add $CATALOG_HRN nodecardinality-count nodecardinality-count --versioned --summary "nodecardinality count" \
            --description "nodecardinality count" --partitioning heretile:12 \
            --content-type application/json --scope $PROJECT_HRN
olp catalog layer add $CATALOG_HRN state state --versioned --summary "state" --description "state" \
            --partitioning Generic --content-type application/octet-stream \
            --scope $PROJECT_HRN

Note::

If a billing tag is required in your realm, use the --billing-tags: "YOUR_BILLING_TAG" parameter.

  1. Use the olp project resource link command to link the HERE Map Content catalog to your project:
olp project resource link $PROJECT_HRN $CATALOG_RIB

Configure the Compiler

From the SDK examples directory, open the data-processing/scala/stateful-nodecardinality-extraction project in your Integrated Development Environment (IDE).

The config/here/pipeline-config.conf (for the HERE platform environment) file contains the permanent configuration of the data sources for the compiler.

Pick the file that corresponds to your platform environemnt. For example, the pipeline configuration for the HERE platform environment looks like:

pipeline.config {
  output-catalog {hrn = "YOUR_OUTPUT_CATALOG_HRN"}
  input-catalogs {
    rib {hrn = "hrn:here:data::olp-here:rib-2"}
  }
}

Replace YOUR_OUTPUT_CATALOG_HRN with the HRN of your nodecardinality catalog. To find the HRN, in the HERE platform portal, navigate to your catalog. The HRN is displayed in the upper left corner of the page.

The config/here/pipeline-job-first.conf and config/here/pipeline-second.conf files contain the compiler's run configuration and point to two different versions of the HERE Map Content Catalog.

To find the latest version of the HERE Map Content catalog, in the HERE platform portal, navigate to the HERE Map Content catalog, and view the current catalog's version in the Catalog info section.

The remainder of the configuration is specified in the application.conf file that can be found in the src/main/resources directory of the compiler project. However, you do not have to modify it unless you want to change the behavior of the compiler.

Generate a Fat JAR file

Run the sbt assembly command in the stateful-nodecardinality-extraction directory to generate a fat JAR file to deploy the compiler to a pipeline.

sbt assembly

Deploy the Compiler to a Pipeline

Once the previous command is finished, your JAR is then available at the target directory, and you can upload it using the HERE pipeline UI or the OLP CLI.

You can use the OLP CLI to create pipeline components and activate the pipeline version with the following commands:

  1. Create pipeline components:

For this example, a bounding box filter is provided by --runtime-config parameter to limit the number of partitions to be processed. This speeds up the compilation process. In this example, we use a bounding box around the cities of Berlin. You can edit the bounding box coordinates to compile a different partition of HERE Map Content. Make sure you update the layer coverage to reflect the different geographical region.

olp pipeline create $COMPONENT_NAME_Pipeline --scope $PROJECT_HRN
olp pipeline template create $COMPONENT_NAME_Template batch-4.0 $PATH_TO_JAR \
                com.here.platform.data.processing.example.scala.feedback.Main \
                --workers=4 --worker-units=3 --supervisor-units=2 --input-catalog-ids=rib \
                --scope $PROJECT_HRN
olp pipeline version create $COMPONENT_NAME_version $PIPELINE_ID $PIPELINE_TEMPLATE_ID \
                "$PATH_TO_CONFIG_FOLDER/pipeline-config.conf" \
                --runtime-config  here.platform.data-processing.executors.partitionKeyFilters.0.className=BoundingBoxFilter \
                                  here.platform.data-processing.executors.partitionKeyFilters.0.param.boundingBox.north=52.67551 \
                                  here.platform.data-processing.executors.partitionKeyFilters.0.param.boundingBox.south=52.338261 \
                                  here.platform.data-processing.executors.partitionKeyFilters.0.param.boundingBox.east=13.76116 \
                                  here.platform.data-processing.executors.partitionKeyFilters.0.param.boundingBox.west=13.08835 \
                --scope $PROJECT_HRN
  1. Activate the pipeline version:
olp pipeline version activate $PIPELINE_ID $PIPELINE_VERSION_ID \
                --input-catalogs "$PATH_TO_CONFIG_FOLDER/pipeline-job-first.conf" \
                --scope $PROJECT_HRN

You do not have to specify the input catalog's version, unless you want to. The latest version will be automatically used.

In the HERE platform portal, navigate to your pipeline to see its status.

Verify the Output

In the HERE platform portal, select the Data tab and find your catalog.

  1. Open the nodecardinality-count layer and select the Inspect tab. Verify that partitions with the JSON data are present, and you can view this data by selecting a partition.
  2. Select any partition to look at its content. You should see the nodes with their cardinality, for example here:cm:node:1002845384: 3, where here:cm:node:1002845384 is the node, and 3 - cardinality. The field updatesCount will be 1 for those partitions that did not change during the second run. It will be 2 otherwise. The JSON output of the compiler should be displayed on the right side as follows:

Example results