hadoop_indexing_task.json

{
  "type" : "index_hadoop",
  "spec" : {
    "dataSchema" : {
      "dataSource" : "gwiq",
      "parser" : {
        "type" : "hadoopyString",
        "parseSpec" : {
          "format" : "json",
          "timestampSpec" : {
            "column" : "time",
            "format" : "auto"
          },
          "dimensionsSpec" : {
            "dimensions": [],
            "dimensionExclusions" : ["time", "gwid", "price"],
            "spatialDimensions" : []
          }
        }
      },
      "metricsSpec" : [
        {
          "type" : "count",
          "name" : "count"
        },
        {
          "type": "hyperUnique",
          "fieldName": "gwid",
          "name": "gwidHll"
        },
        {
          "type" : "doubleSum",
          "name" : "priceSum",
          "fieldName" : "price"
        }
      ],
      "granularitySpec" : {
        "type" : "uniform",
        "segmentGranularity" : "HOUR",
        "queryGranularity" : "all",
        "intervals" : [ "2015-01-01T00Z/2015-01-01T04Z" ]
      }
    },
    "ioConfig" : {
      "type" : "hadoop",
      "inputSpec" : {
        "type" : "granularity",
        "dataGranularity" : "HOUR",
        "inputPath" : "/var/data/druid/events/",
        "filePattern" : ".*",
        "pathFormat" : "yyyy/MM/dd/HH"
      }
    },
    "tuningConfig" : {
      "type": "hadoop"
    }
  }
}