You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@carbondata.apache.org by "Chetan Bhat (Jira)" <ji...@apache.org> on 2021/06/07 05:41:00 UTC

[jira] [Created] (CARBONDATA-4200) query with AND filter and query on partition table not hitting SI from presto cli

Chetan Bhat created CARBONDATA-4200:
---------------------------------------

             Summary: query with AND filter and query on partition table not hitting SI from presto cli
                 Key: CARBONDATA-4200
                 URL: https://issues.apache.org/jira/browse/CARBONDATA-4200
             Project: CarbonData
          Issue Type: Bug
          Components: presto-integration
    Affects Versions: 2.1.1
         Environment: Spark 2.4.5 , Presto 316/333
            Reporter: Chetan Bhat


[Steps] :-

Issue 1 : - query with AND filter not hitting SI from presto cli.

Queries executed from spark sql/beeline - 

drop table if exists uniqdata;

CREATE TABLE uniqdata (CUST_ID int,CUST_NAME String,ACTIVE_EMUI_VERSION string, DOB timestamp, DOJ timestamp, BIGINT_COLUMN1 bigint,BIGINT_COLUMN2 bigint,DECIMAL_COLUMN1 decimal(30,10), DECIMAL_COLUMN2 decimal(36,36),Double_COLUMN1 double, Double_COLUMN2 double,INTEGER_COLUMN1 int) stored as carbondata ;

create index uniqdata_index on table uniqdata(cust_name) as 'carbondata' properties ('sort_scope'='global_sort', 'Global_sort_partitions'='3');

create index uniqdata_index1 on table uniqdata(ACTIVE_EMUI_VERSION) as 'carbondata' properties ('sort_scope'='global_sort', 'Global_sort_partitions'='3');

LOAD DATA INPATH 'hdfs://hacluster/chetan/2000_UniqData.csv' into table uniqdata OPTIONS('DELIMITER'=',' , 'QUOTECHAR'='"','BAD_RECORDS_ACTION'='FORCE','FILEHEADER'='CUST_ID,CUST_NAME,ACTIVE_EMUI_VERSION,DOB,DOJ,BIGINT_COLUMN1,BIGINT_COLUMN2,DECIMAL_COLUMN1,DECIMAL_COLUMN2,Double_COLUMN1,Double_COLUMN2,INTEGER_COLUMN1');

Queries executed from presto cli - 

select * from uniqdata where cust_name='CUST_NAME_00002' AND ACTIVE_EMUI_VERSION='ACTIVE_EMUI_VERSION_00002'; show metacache on table uniqdata;

 

Issue 2 - query on partition table does not hit SI

Queries executed from spark sql/beeline - 

drop table carbon_test;

create table carbon_test(id int,name string)PARTITIONED BY(record_date int) stored as carbondata TBLPROPERTIES('SORT_COLUMNS'='id','SORT_SCOPE'='NO_SORT');

create index uniq1 on table carbon_test (name) as 'carbondata';

insert into table carbon_test partition(record_date) select 1,'kim',unix_timestamp('2018-02-05','yyyy-MM-dd') as record_date ;

Queries executed from presto cli - 

select * from carbon_test where name='kim'; show metacache on table carbon_test;



--
This message was sent by Atlassian Jira
(v8.3.4#803005)