lineitem.task.json

{
  "type" : "index_hadoop",
  "hadoopCoordinates":"org.apache.hadoop:hadoop-core:0.20.205-emr",
  "config": {
    "dataSource" : "tpch_lineitem",
    "timestampSpec" : {
      "column": "l_shipdate",
      "format": "yyyy-MM-dd"
    },
    "dataSpec": {
      "format": "tsv",
      "delimiter": "|",
      "columns": [
        "l_orderkey",
        "l_partkey",
        "l_suppkey",
        "l_linenumber",
        "l_quantity",
        "l_extendedprice",
        "l_discount",
        "l_tax",
        "l_returnflag",
        "l_linestatus",
        "l_shipdate",
        "l_commitdate",
        "l_receiptdate",
        "l_shipinstruct",
        "l_shipmode",
        "l_comment"
      ],
      "dimensions" : [
        "l_orderkey",
        "l_partkey",
        "l_suppkey",
        "l_linenumber",
        "l_returnflag",
        "l_linestatus",
        "l_shipdate",
        "l_commitdate",
        "l_receiptdate",
        "l_shipinstruct",
        "l_shipmode",
        "l_comment"
      ]
    },
    "granularitySpec": {
      "type": "uniform",
      "intervals": [
        "1980/2020"
      ],
      "gran": "month"
    },
    "partitionsSpec": {
      "type": "random",
      "targetPartitionSize": 5000000
    },
    "pathSpec": {
      "type": "static",
      "paths": "s3:\/\/path\/to\/tpch\/data\/lineitem.tbl.*.gz"
    },
    "rollupSpec": {
      "aggs": [
        {
          "type": "count",
          "name": "count"
        },
        {
          "type": "longSum",
          "fieldName": "L_QUANTITY",
          "name": "L_QUANTITY"
        },
        {
          "type": "doubleSum",
          "fieldName": "L_EXTENDEDPRICE",
          "name": "L_EXTENDEDPRICE"
        },
        {
          "type": "doubleSum",
          "fieldName": "L_DISCOUNT",
          "name": "L_DISCOUNT"
        },
        {
          "type": "doubleSum",
          "fieldName": "L_TAX",
          "name": "L_TAX"
        }
      ],
      "rollupGranularity": "day"
    }
  }
}