[jira] [Created] (CARBONDATA-786) Data mismatch if the data data is loaded across blocklet groups

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[jira] [Created] (CARBONDATA-786) Data mismatch if the data data is loaded across blocklet groups

Akash R Nilugal (Jira)
Ravindra Pesala created CARBONDATA-786:
------------------------------------------

             Summary: Data mismatch if the data data is loaded across blocklet groups
                 Key: CARBONDATA-786
                 URL: https://issues.apache.org/jira/browse/CARBONDATA-786
             Project: CarbonData
          Issue Type: Bug
            Reporter: Ravindra Pesala


Data mismatch if the data data is loaded across blocklet groups and filter applied on second column onwards.

Follow testcase

{code}

CarbonProperties.getInstance()
      .addProperty("carbon.blockletgroup.size.in.mb", "16")
      .addProperty("carbon.enable.vector.reader", "true")
      .addProperty("enable.unsafe.sort", "true")

    val rdd = sqlContext.sparkContext
      .parallelize(1 to 1200000, 4)
      .map { x =>
        ("city" + x % 8, "country" + x % 1103, "planet" + x % 10007, x.toString,
          (x % 16).toShort, x / 2, (x << 1).toLong, x.toDouble / 13, x.toDouble / 11)
      }.map { x =>
      Row(x._1, x._2, x._3, x._4, x._5, x._6, x._7, x._8, x._9)
    }

    val schema = StructType(
      Seq(
        StructField("city", StringType, nullable = false),
        StructField("country", StringType, nullable = false),
        StructField("planet", StringType, nullable = false),
        StructField("id", StringType, nullable = false),
        StructField("m1", ShortType, nullable = false),
        StructField("m2", IntegerType, nullable = false),
        StructField("m3", LongType, nullable = false),
        StructField("m4", DoubleType, nullable = false),
        StructField("m5", DoubleType, nullable = false)
      )
    )

    val input = sqlContext.createDataFrame(rdd, schema)
    sql(s"drop table if exists testBigData")
    input.write
      .format("carbondata")
      .option("tableName", "testBigData")
      .option("tempCSV", "false")
      .option("single_pass", "true")
      .option("dictionary_exclude", "id") // id is high cardinality column
      .mode(SaveMode.Overwrite)
      .save()
sql(s"select city, sum(m1) from testBigData " +
          s"where country='country12' group by city order by city").show()
{code}

The above code supposed return following data, but not returning it.
{code}
+-----+-------+
| city|sum(m1)|
+-----+-------+
|city0|    544|
|city1|    680|
|city2|    816|
|city3|    952|
|city4|   1088|
|city5|   1224|
|city6|   1360|
|city7|   1496|
+-----+-------+
{code}



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)