mirror of
https://github.com/apache/impala.git
synced 2026-01-07 18:02:33 -05:00
6d87fe090c
Based on Google's HyperLogLog++ paper. Uses a bias correcting interpolation as a sub algorithm for Hll estimates within a specific range. Change-Id: If4fe692b4308f6a57aea6167e9bc00db11eaaab9 Reviewed-on: http://gerrit.cloudera.org:8080/415 Tested-by: Internal Jenkins Reviewed-by: Henry Robinson <henry@cloudera.com>
143 lines
4.2 KiB
Plaintext
143 lines
4.2 KiB
Plaintext
====
|
|
---- QUERY
|
|
# test computing stats on an HBase table
|
|
create table compute_stats_db_hbase.alltypessmall_hbase
|
|
like alltypessmall;
|
|
====
|
|
---- QUERY
|
|
compute stats compute_stats_db_hbase.alltypessmall_hbase
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 13 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats compute_stats_db_hbase.alltypessmall_hbase
|
|
---- LABELS
|
|
REGION LOCATION, START ROWKEY, EST. #ROWS, SIZE
|
|
---- RESULTS: VERIFY_IS_EQUAL
|
|
regex:.+,'',regex:.+,regex:.+KB
|
|
regex:.+,'1',regex:.+,regex:.+KB
|
|
regex:.+,'3',regex:.+,regex:.+KB
|
|
regex:.+,'5',regex:.+,regex:.+KB
|
|
regex:.+,'7',regex:.+,regex:.+KB
|
|
regex:.+,'9',regex:.+,regex:.+KB
|
|
'Total','',regex:.+,regex:.+KB
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats compute_stats_db_hbase.alltypessmall_hbase
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE
|
|
---- RESULTS
|
|
'id','INT',99,-1,4,4
|
|
'bigint_col','BIGINT',10,-1,8,8
|
|
'bool_col','BOOLEAN',2,-1,1,1
|
|
'date_string_col','STRING',12,-1,8,8
|
|
'double_col','DOUBLE',10,-1,8,8
|
|
'float_col','FLOAT',10,-1,4,4
|
|
'int_col','INT',10,-1,4,4
|
|
'month','INT',4,-1,4,4
|
|
'smallint_col','SMALLINT',10,-1,2,2
|
|
'string_col','STRING',10,-1,1,1
|
|
'timestamp_col','TIMESTAMP',101,-1,16,16
|
|
'tinyint_col','TINYINT',10,-1,1,1
|
|
'year','INT',1,-1,4,4
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, INT, DOUBLE
|
|
====
|
|
---- QUERY
|
|
# test computing stats on an binary HBase table
|
|
create table compute_stats_db_hbase.alltypessmall_hbase_bin
|
|
like alltypessmallbinary;
|
|
====
|
|
---- QUERY
|
|
compute stats compute_stats_db_hbase.alltypessmall_hbase_bin
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 13 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY: VERIFY_IS_EQUAL
|
|
show table stats compute_stats_db_hbase.alltypessmall_hbase_bin
|
|
---- LABELS
|
|
REGION LOCATION, START ROWKEY, EST. #ROWS, SIZE
|
|
---- RESULTS
|
|
regex:.+,'',regex:.+,regex:.+
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats compute_stats_db_hbase.alltypessmall_hbase_bin
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE
|
|
---- RESULTS
|
|
'id','INT',99,-1,4,4
|
|
'bigint_col','BIGINT',10,-1,8,8
|
|
'bool_col','BOOLEAN',2,-1,1,1
|
|
'date_string_col','STRING',12,-1,8,8
|
|
'double_col','DOUBLE',10,-1,8,8
|
|
'float_col','FLOAT',10,-1,4,4
|
|
'int_col','INT',10,-1,4,4
|
|
'month','INT',4,-1,4,4
|
|
'smallint_col','SMALLINT',10,-1,2,2
|
|
'string_col','STRING',10,-1,1,1
|
|
'timestamp_col','TIMESTAMP',101,-1,16,16
|
|
'tinyint_col','TINYINT',10,-1,1,1
|
|
'year','INT',1,-1,4,4
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, INT, DOUBLE
|
|
====
|
|
---- QUERY
|
|
# IMP-1227: Test computing stats on an HBase table that has a
|
|
# complex-typed column that Impala does not yet support.
|
|
create table compute_stats_db_hbase.allcomplextypes
|
|
like allcomplextypes
|
|
====
|
|
---- QUERY
|
|
compute stats compute_stats_db_hbase.allcomplextypes
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 3 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY: VERIFY_IS_EQUAL
|
|
show table stats compute_stats_db_hbase.allcomplextypes
|
|
---- LABELS
|
|
REGION LOCATION, START ROWKEY, EST. #ROWS, SIZE
|
|
---- RESULTS
|
|
regex:.+,'',regex:.+,regex:.+
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats compute_stats_db_hbase.allcomplextypes
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE
|
|
---- RESULTS
|
|
'id','INT',0,-1,4,4
|
|
'array_array_col','ARRAY<ARRAY<INT>>',-1,-1,-1,-1
|
|
'array_map_col','MAP<STRING,ARRAY<INT>>',-1,-1,-1,-1
|
|
'complex_nested_struct_col','STRUCT<f1:INT,f2:ARRAY<STRUCT<f11:BIGINT,f12:MAP<STRING,STRUCT<f21:BIGINT>>>>>',-1,-1,-1,-1
|
|
'complex_struct_col','STRUCT<f1:INT,f2:ARRAY<INT>,f3:MAP<STRING,INT>>',-1,-1,-1,-1
|
|
'int_array_col','ARRAY<INT>',-1,-1,-1,-1
|
|
'int_map_col','MAP<STRING,INT>',-1,-1,-1,-1
|
|
'int_struct_col','STRUCT<f1:INT,f2:INT>',-1,-1,-1,-1
|
|
'map_array_col','ARRAY<MAP<STRING,INT>>',-1,-1,-1,-1
|
|
'map_map_col','MAP<STRING,MAP<STRING,INT>>',-1,-1,-1,-1
|
|
'month','INT',0,-1,4,4
|
|
'nested_struct_col','STRUCT<f1:INT,f2:STRUCT<f11:BIGINT,f12:STRUCT<f21:BIGINT>>>',-1,-1,-1,-1
|
|
'struct_array_col','ARRAY<STRUCT<f1:BIGINT,f2:STRING>>',-1,-1,-1,-1
|
|
'struct_map_col','MAP<STRING,STRUCT<f1:BIGINT,f2:STRING>>',-1,-1,-1,-1
|
|
'year','INT',0,-1,4,4
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, INT, DOUBLE
|
|
====
|
|
---- QUERY
|
|
compute incremental stats alltypes;
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 13 column(s).'
|
|
---- TYPES
|
|
STRING
|