// Add steps as necessary for accessing the software, post-configuration, and testing. Don’t include full usage instructions for your software, but add links to your product documentation for that information. //Should any sections not be applicable, remove them :xrefstyle: short == MRASCo data adapter option This Quick Start features a MRASCo data adapter option to transform MRASCo data flow files to MDA landing zone format. To deploy this feature, choose `mrasco` for the `DeploySpecialAdapters` parameter during deployment. The MRASCo data adapter option deploys the following resources: * An S3 bucket to store MRASCo data files. The S3 bucket is configured with an agent to invoke an AWS Lambda function when files are uploaded. * An AWS Lambda function to retrieve MRASCo data files from the S3 bucket and transform them to MDA landing zone format. == Postdeployment steps To see the capabilities of the QuickStart, follow the steps in this section to use the dataset of meter reads from the City of London between 2011 and 2014. The link:#_customize_this_quick_start[Customize this Quick Start] section explains how to edit the raw-data ETL script to work with your own meter data. === Download demo dataset NOTE: To use the demo dataset, you must set the `LandingzoneTransformer` parameter to `London` during Quick Start deployment. Download the sample dataset from https://data.london.gov.uk/dataset/smartmeter-energy-use-data-in-london-households[SmartMeter Energy Consumption Data in London Households^]. The file can be downloaded to your local machine, unzipped, and then uploaded to Amazon S3. However, given the size of the unzipped file (~11 GB), it's faster to run an EC2 instance with sufficient permissions to write to Amazon S3, download the file to the EC2 instance, unzip it, and upload the data from there. https://data.london.gov.uk/download/smartmeter-energy-use-data-in-london-households/3527bf39-d93e-4071-8451-df2ade1ea4f2/Power-Networks-LCL-June2015(withAcornGps).zip[Download sample meter data^] Mirror: https://www.kaggle.com/jeanmidev/smart-meters-in-london?select=halfhourly_dataset.zip[Download from kaggle] === Upload the dataset to the landing-zone S3 bucket You must upload the sample dataset to the landing-zone (raw-data) S3 bucket. The landing zone is the starting point for the AWS Glue workflow. Files that are placed in this S3 bucket are processed by the ETL pipeline. Furthermore, the AWS Glue extract, transform, and load (ETL) workflow tracks which files have been processed and which have not. To upload the sample dataset to the landing-zone S3 bucket, follow these steps: . Choose the S3 bucket with the name that contains `landingzones`. This is the starting point for the ETL process. + [#bucket_layout] .S3 bucket layout [link=images/1_bucket_layout.png] image::../images/1_bucket_layout.png[Bucket Layout] . Upload the London meter data to the landing-zone bucket. + [#upload_demo_dataset] .Uploading the demo dataset [link=images/2_upload_demo_data_set.png] image::../images/2_upload_demo_data_set.png[Upload demo dataset] . Verify that the directory contains one file with meter data and no subdirectories. + [#uploaded_demo_dataset] .The uploaded demo dataset [link=images/3_upload_demo_data_set.png] image::../images/3_upload_demo_data_set.png[Uploaded demo dataset] WARNING: The ETL workflow fails if the S3 bucket contains subdirectories. //[TODO EC2 command line steps] === (Optional) Prepare weather data for use in training and forecasting :xrefstyle: short . https://www.kaggle.com/jeanmidev/smart-meters-in-london?select=weather_hourly_darksky.csv[Download^] the sample weather dataset. . Upload the dataset to a new S3 bucket. The bucket can have any name. You must reference the bucket in a SQL statement in the next step. . Create a weather table in the target AWS Glue database using the SQL statement in <>. Replace `` in the code with the location of the S3 bucket with the weather dataset. + [#create_weather_table] .Creating the weather table [link=images/4_create_weather_table.png] image::../images/4_create_weather_table.png[Create weather table via Athena] + ```sql CREATE EXTERNAL TABLE weather( visibility double, windbearing bigint, temperature double, time timestamp, dewpoint double, pressure double, apparenttemperature double, windspeed double, preciptype string, icon string, humidity double, summary string ) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' LOCATION 's3:///' TBLPROPERTIES ( 'has_encrypted_data'='false', 'skip.header.line.count'='1'); ``` If you have another weather dataset you want to use, verify that it contains at least the four columns shown in <>. [cols="1,1,1,1", options="header"] [#weather_data_schema] .Weather data schema |=== |Field |Type |Mandatory |Format |time|string|yes| `yyyy-MM-dd HH:mm:ss` |temperature| double|yes| |apparenttemperature|double|yes| |humidity|double|yes| |=== NOTE: To enable the use of weather data, set the `WithWeather` parameter to `1` during Quick Start deployment. === (Optional) Prepare geolocation data This Quick Start requires geolocation data to be uploaded to the `geodata` directory in the business-zone S3 bucket. The name of this bucket appears in the *Outputs* section in the AWS CloudFormation stack. The S3 path of geodata location should be `s3:///geodata/`. The data should be in a CSV file format with the following structure, where all three fields (meter id, latitude, and longitude) are mandatory as comma-separated values. Sample data is located in quickstart-aws-utility-meter-data-analytics-platform/assets/data/meter-geo-data.csv [cols="1,1", options="header"] .Geolocation data schema |=== |Field Name |Field Type |meter id |string |latitude |string |longitude |string |=== `meter id`, `latitude`, and `longitude` are mandatory. If data are missing, API requests for meter-outage information returns an error. The grafana dashboards require a fuller set of geolocation data. This data is located in quickstart-aws-utility-meter-data-analytics-platform/assets/data/Full-Geodata.csv. To add this data, upload to the same s3 bucket as your weather data and run the following in Athena. ```sql CREATE EXTERNAL TABLE `full_geo`( `id` string, `lat` double, `long` double, `streetnumber` string, `streetname` string, `municipalitysubdivision` string, `municipality` string, `countrysecondarysubdivision` string, `countrysubdivision` string, `country` string, `countrycode` string, `countrycodeiso3` string, `postalcode` string, `order` int, `daily_outages` int, `weekly_outages` int, `monthly_outages` int) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' STORED AS INPUTFORMAT 'org.apache.hadoop.mapred.TextInputFormat' OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat' LOCATION 's3:///' TBLPROPERTIES ( 'has_encrypted_data'='false', 'skip.header.line.count'='1'); ``` === Start the AWS Glue ETL and ML training pipeline manually By default, the pipeline is invoked each day at 9:00 a.m. to process newly arrived data in the landing-zone S3 bucket. To start the ETL pipeline manually, do the following: . Open the AWS Glue console. Choose *Workflows*. + [#glue_console] [link=images/4_start_etl_workflow.png] .Viewing AWS Glue workflows image::../images/4_start_etl_workflow.png[AWS Glue console] . Select the workflow. . On the *Actions* menu, choose *Run*. + [#start_etl_workflow] .Start the ETL workflow [link=images/5_start_etl_workflow.png] image::../images/5_start_etl_workflow.png[Start ETL workflow] . In the *History* tab, ensure that the workflow run status is *Running*. + [#running_workflow] .Running the workflow [link=images/6_start_etl_workflow.png] image::../images/6_start_etl_workflow.png[Running workflow] == Customize this Quick Start You can customize this Quick Start for use with your own data format. To do that, adjust the first AWS Glue job to map the incoming data to the internal meter-data-lake schema. The following steps describe how. The first AWS Glue job in the ETL workflow transforms the raw data in the landing-zone S3 bucket to clean data in the clean-zone S3 bucket. This step also maps the inbound data to the internal data schema, which is used by the rest of the steps in the AWS ETL workflow and the ML state machines. To update the data mapping, you can edit the AWS Glue job directly in the web editor. . Open the AWS Glue console. Choose **Jobs**. + [#glue_job_console] .Viewing AWS Glue jobs [link=images/1_edit_etl_job.png] image::../images/1_edit_etl_job.png[AWS Glue Job console] . Select the ETL job `transform_raw_to_clean-`. + [#edit_etl_job] .Selecting the ETL job [link=images/2_edit_etl_job.png] image::../images/2_edit_etl_job.png[Select the ETL job] . Choose *Action, Edit script*. Edit the input mapping in the script editor. + [#open_editor] .Opening the script editor [link=images/3_open_editor.png] image::../images/3_open_editor.png[Open the script editor] . To adopt a different input format, edit the `ApplyMapping` call. The internal model works with the following schema: + [cols="1,1,1,1", options="header"] .Schema |=== |Field |Type |Mandatory |Format |meter_id| String| yes| |reading_value| double| yes|0.000 |reading_type| String| yes|AGG\|INT |reading_date_time| Timestamp| yes|yyyy-MM-dd HH:mm:ss.SSS |date_str| String|yes| yyyyMMdd |obis_code| String|no | |week_of_year| int|no | |month| int|no | |year| int|no | |hour| int|no | |minute| int|no | |=== include::./ml_pipeline.adoc[] include::./grafana.adoc[] == Data schema and API I/O format include::./data_format.adoc[] == Best practices Use https://aws.amazon.com/s3/glacier/[Amazon S3 Glacier] to archive meter data from raw, clean, and business S3 buckets for long-term storage and cost savings.