Ingest in-situ data (in json) to AWS S3 as parquet object files.

Follow this guide to deploy SDAP In-Situ to AWS cloud.

• Assumption: K8s is successfully deployed

• Download this repo

• (optional) create different python3.6 environment

• install dependencies

    python3 setup.py install
  

• setup AWS tokens

    export AWS_ACCESS_KEY_ID=xxx
    export AWS_SECRET_ACCESS_KEY=xxx
    export AWS_SESSION_TOKEN=really.long.token
    export AWS_REGION=us-west-2
  

  • alternatively the default profile under ~/.aws/credentials can be setup as well

• setup current directory to PYTHONPATH

    PYTHONPATH="${PYTHONPATH}:/absolute/path/to/current/dir/"
  

• run the script:

    python3 -m parquet_cli.ingest_s3 --help
  

  • sample script:

      python3 -m parquet_cli.ingest_s3 \
        --LOG_LEVEL 30 \
        --CDMS_DOMAIN https://doms.jpl.nasa.gov/insitu  \
        --CDMS_BEARER_TOKEN Mock-CDMS-Flask-Token  \
        --PARQUET_META_TBL_NAME cdms_parquet_meta_dev_v1  \
        --BUCKET_NAME cdms-dev-ncar-in-situ-stage  \
        --KEY_PREFIX cdms_icoads_2017-01-01.json
    

• to replace parquet file partially

https://stackoverflow.com/questions/38487667/overwrite-specific-partitions-in-spark-dataframe-write-method?noredirect=1&lq=1
> Finally! This is now a feature in Spark 2.3.0: SPARK-20236
> To use it, you need to set the spark.sql.sources.partitionOverwriteMode setting to dynamic, the dataset needs to be partitioned, and the write mode overwrite. Example:
> https://stackoverflow.com/questions/50006526/overwrite-only-some-partitions-in-a-partitioned-spark-dataset

spark.conf.set("spark.sql.sources.partitionOverwriteMode","dynamic")
data.toDF().write.mode("overwrite").format("parquet").partitionBy("date", "name").save("s3://path/to/somewhere")