[CARBONDATA-2731] Timeseries datamap queries should fetch data from the Timeseries datamap.

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 1.4.1
• Component/s: data-query
• Environment: Spark 2.2
• Assignee: Unassigned
• Reporter: Prasanna Ravichandran
• Created: 11/Jul/18 12:38
• Updated: 11/Jul/18 12:41

Description

While creation of the timeseries datamap, queries to which it would apply would also be defined. SO when the user uses that same query after creation of TS datamap, then that query should fetch the data from the TS datamap created. 

Test queries:

create table brinjal (imei string,AMSize string,channelsId string,ActiveCountry string, Activecity string,gamePointId double,deviceInformationId double,productionDate Timestamp,deliveryDate timestamp,deliverycharge double) STORED BY 'org.apache.carbondata.format' TBLPROPERTIES('table_blocksize'='1');
LOAD DATA INPATH 'hdfs://hacluster/user/prasanna/vardhandaterestruct.csv' INTO TABLE brinjal OPTIONS('DELIMITER'=',', 'QUOTECHAR'= '"','BAD_RECORDS_ACTION'='FORCE','FILEHEADER'= 'imei,deviceInformationId,AMSize,channelsId,ActiveCountry,Activecity,gamePointId,productionDate,deliveryDate,deliverycharge');

CREATE DATAMAP agg0_time ON TABLE brinjal USING 'timeSeries' DMPROPERTIES ('EVENT_TIME'='productionDate','SECOND_GRANULARITY'='1') AS SELECT productionDate, SUM(imei) FROM brinjal GROUP BY productionDate;

explain SELECT productionDate, SUM(imei) FROM brinjal GROUP BY productionDate; 

0: jdbc:hive2://10.18.98.136:23040/default> show datamap on table brinjal;
--------------------------++++----------------------------------------------------------------------------------

DataMapName

ClassName

Associated Table

DataMap Properties

--------------------------++++----------------------------------------------------------------------------------

agg0_time	timeSeries	rp.brinjal_agg0_time	'event_time'='productionDate', 'second_granularity'='1'
sensor	preaggregate	rp.brinjal_sensor

--------------------------++++----------------------------------------------------------------------------------
2 rows selected (0.042 seconds)
0: jdbc:hive2://10.18.98.136:23040/default> explain SELECT productionDate, SUM(imei) FROM brinjal GROUP BY productionDate;
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+

plan

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+

== CarbonData Profiler == Table Scan on brinjal total blocklets: 1 filter: none pruned by Main DataMap skipped blocklets: 0

== Physical Plan == *HashAggregate(keys=productionDate#155228, functions=sum(cast(imei#155221 as double))) +- Exchange hashpartitioning(productionDate#155228, 200) +- *HashAggregate(keys=productionDate#155228, functions=partial_sum(cast(imei#155221 as double))) +- BatchedScan CarbonDatasourceHadoopRelation [ Database name :rp, *Table name :brinjal, Schema :Some(StructType(StructField(imei,StringType,true), StructField(amsize,StringType,true), StructField(channelsid,StringType,true), StructField(activecountry,StringType,true), StructField(activecity,StringType,true), StructField(gamepointid,DoubleType,true), StructField(deviceinformationid,DoubleType,true), StructField(productiondate,TimestampType,true), StructField(deliverydate,TimestampType,true), StructField(deliverycharge,DoubleType,true))) ] rp.brinjalimei#155221,productiondate#155228

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
2 rows selected (0.155 seconds)

 

Expected result: Timeseries datamap queries should fetch data from the Timeseries datamap.

Actual result: Timeseries datamap queries are fetching data from the main table.