mirror of
https://github.com/apache/impala.git
synced 2026-01-04 09:00:56 -05:00
ed7f5ebf53
Previously Impala disallowed LOAD DATA and INSERT on S3. This patch functionally enables LOAD DATA and INSERT on S3 without making major changes for the sake of improving performance over S3. This patch also enables both INSERT and LOAD DATA between file systems. S3 does not support the rename operation, so the staged files in S3 are copied instead of renamed, which contributes to the slow performance on S3. The FinalizeSuccessfulInsert() function now does not make any underlying assumptions of the filesystem it is on and works across all supported filesystems. This is done by adding a full URI field to the base directory for a partition in the TInsertPartitionStatus. Also, the HdfsOp class now does not assume a single filesystem and gets connections to the filesystems based on the URI of the file it is operating on. Added a python S3 client called 'boto3' to access S3 from the python tests. A new class called S3Client is introduced which creates wrappers around the boto3 functions and have the same function signatures as PyWebHdfsClient by deriving from a base abstract class BaseFileSystem so that they can be interchangeably through a 'generic_client'. test_load.py is refactored to use this generic client. The ImpalaTestSuite setup creates a client according to the TARGET_FILESYSTEM environment variable and assigns it to the 'generic_client'. P.S: Currently, the test_load.py runs 4x slower on S3 than on HDFS. Performance needs to be improved in future patches. INSERT performance is slower than on HDFS too. This is mainly because of an extra copy that happens between staging and the final location of a file. However, larger INSERTs come closer to HDFS permformance than smaller inserts. ACLs are not taken care of for S3 in this patch. It is something that still needs to be discussed before implementing. Change-Id: I94e15ad67752dce21c9b7c1dced6e114905a942d Reviewed-on: http://gerrit.cloudera.org:8080/2574 Reviewed-by: Sailesh Mukil <sailesh@cloudera.com> Tested-by: Internal Jenkins
168 lines
6.9 KiB
Python
168 lines
6.9 KiB
Python
# Copyright (c) 2012 Cloudera, Inc. All rights reserved.
|
|
# Targeted tests for Impala joins
|
|
#
|
|
import logging
|
|
import os
|
|
import pytest
|
|
from copy import copy
|
|
from tests.common.test_vector import *
|
|
from tests.common.impala_test_suite import *
|
|
from tests.common.skip import SkipIf, SkipIfS3, SkipIfIsilon, SkipIfOldAggsJoins, SkipIfLocal
|
|
|
|
class TestJoinQueries(ImpalaTestSuite):
|
|
BATCH_SIZES = [0, 1]
|
|
|
|
@classmethod
|
|
def get_workload(cls):
|
|
return 'functional-query'
|
|
|
|
@classmethod
|
|
def add_test_dimensions(cls):
|
|
super(TestJoinQueries, cls).add_test_dimensions()
|
|
cls.TestMatrix.add_dimension(
|
|
TestDimension('batch_size', *TestJoinQueries.BATCH_SIZES))
|
|
# TODO: Look into splitting up join tests to accomodate hbase.
|
|
# Joins with hbase tables produce drastically different results.
|
|
cls.TestMatrix.add_constraint(lambda v:\
|
|
v.get_value('table_format').file_format in ['parquet'])
|
|
|
|
if cls.exploration_strategy() != 'exhaustive':
|
|
# Cut down on execution time when not running in exhaustive mode.
|
|
cls.TestMatrix.add_constraint(lambda v: v.get_value('batch_size') != 1)
|
|
|
|
def test_basic_joins(self, vector):
|
|
new_vector = copy(vector)
|
|
new_vector.get_value('exec_option')['batch_size'] = vector.get_value('batch_size')
|
|
self.run_test_case('QueryTest/joins', new_vector)
|
|
|
|
@SkipIfOldAggsJoins.unsupported
|
|
def test_partitioned_joins(self, vector):
|
|
self.run_test_case('QueryTest/joins-partitioned', vector)
|
|
|
|
@SkipIfS3.hbase
|
|
@SkipIfIsilon.hbase
|
|
@SkipIf.skip_hbase
|
|
@SkipIfLocal.hbase
|
|
def test_joins_against_hbase(self, vector):
|
|
new_vector = copy(vector)
|
|
new_vector.get_value('exec_option')['batch_size'] = vector.get_value('batch_size')
|
|
self.run_test_case('QueryTest/joins-against-hbase', new_vector)
|
|
|
|
def test_outer_joins(self, vector):
|
|
new_vector = copy(vector)
|
|
new_vector.get_value('exec_option')['batch_size'] = vector.get_value('batch_size')
|
|
self.run_test_case('QueryTest/outer-joins', new_vector)
|
|
|
|
def test_single_node_nested_loop_joins(self, vector):
|
|
# Test the execution of nested-loops joins for join types that can only be
|
|
# executed in a single node (right [outer|semi|anti] and full outer joins).
|
|
new_vector = copy(vector)
|
|
new_vector.get_value('exec_option')['num_nodes'] = 1
|
|
self.run_test_case('QueryTest/single-node-nlj', new_vector)
|
|
|
|
def test_single_node_nested_loop_joins_exhaustive(self, vector):
|
|
if self.exploration_strategy() != 'exhaustive': pytest.skip()
|
|
new_vector = copy(vector)
|
|
new_vector.get_value('exec_option')['num_nodes'] = 1
|
|
self.run_test_case('QueryTest/single-node-nlj-exhaustive', new_vector)
|
|
|
|
class TestTPCHJoinQueries(ImpalaTestSuite):
|
|
# Uses the TPC-H dataset in order to have larger joins. Needed for example to test
|
|
# the repartitioning codepaths.
|
|
|
|
@classmethod
|
|
def get_workload(cls):
|
|
return 'tpch'
|
|
|
|
@classmethod
|
|
def add_test_dimensions(cls):
|
|
super(TestTPCHJoinQueries, cls).add_test_dimensions()
|
|
cls.TestMatrix.add_dimension(
|
|
TestDimension('batch_size', *TestJoinQueries.BATCH_SIZES))
|
|
cls.TestMatrix.add_constraint(lambda v:\
|
|
v.get_value('table_format').file_format in ['parquet'])
|
|
|
|
if cls.exploration_strategy() != 'exhaustive':
|
|
# Cut down on execution time when not running in exhaustive mode.
|
|
cls.TestMatrix.add_constraint(lambda v: v.get_value('batch_size') != 1)
|
|
|
|
@classmethod
|
|
def teardown_class(cls):
|
|
cls.client.execute('set mem_limit = 0');
|
|
super(TestTPCHJoinQueries, cls).teardown_class()
|
|
|
|
def test_outer_joins(self, vector):
|
|
new_vector = copy(vector)
|
|
new_vector.get_value('exec_option')['batch_size'] = vector.get_value('batch_size')
|
|
self.run_test_case('tpch-outer-joins', new_vector)
|
|
|
|
class TestSemiJoinQueries(ImpalaTestSuite):
|
|
@classmethod
|
|
def get_workload(cls):
|
|
return 'functional-query'
|
|
|
|
@classmethod
|
|
def add_test_dimensions(cls):
|
|
super(TestSemiJoinQueries, cls).add_test_dimensions()
|
|
cls.TestMatrix.add_dimension(
|
|
TestDimension('batch_size', *TestJoinQueries.BATCH_SIZES))
|
|
# Joins with hbase tables produce drastically different results.
|
|
cls.TestMatrix.add_constraint(lambda v:\
|
|
v.get_value('table_format').file_format in ['parquet'])
|
|
|
|
if cls.exploration_strategy() != 'exhaustive':
|
|
# Cut down on execution time when not running in exhaustive mode.
|
|
cls.TestMatrix.add_constraint(lambda v: v.get_value('batch_size') != 1)
|
|
|
|
@classmethod
|
|
def setup_class(cls):
|
|
super(TestSemiJoinQueries, cls).setup_class()
|
|
cls.__cleanup_semi_join_tables()
|
|
cls.__load_semi_join_tables()
|
|
|
|
@classmethod
|
|
def teardown_class(cls):
|
|
cls.__cleanup_semi_join_tables()
|
|
super(TestSemiJoinQueries, cls).teardown_class()
|
|
|
|
@classmethod
|
|
def __load_semi_join_tables(cls):
|
|
SEMIJOIN_TABLES = ['functional.SemiJoinTblA', 'functional.SemiJoinTblB']
|
|
# Cleanup, create and load fresh test tables for semi/anti-join tests
|
|
cls.client.execute('create table if not exists '\
|
|
'functional.SemiJoinTblA(a int, b int, c int)')
|
|
cls.client.execute('create table if not exists '\
|
|
'functional.SemiJoinTblB(a int, b int, c int)')
|
|
# loads some values with NULLs in the first table
|
|
cls.client.execute('insert into %s values(1,1,1)' % SEMIJOIN_TABLES[0]);
|
|
cls.client.execute('insert into %s values(1,1,10)' % SEMIJOIN_TABLES[0]);
|
|
cls.client.execute('insert into %s values(1,2,10)' % SEMIJOIN_TABLES[0]);
|
|
cls.client.execute('insert into %s values(1,3,10)' % SEMIJOIN_TABLES[0]);
|
|
cls.client.execute('insert into %s values(NULL,NULL,30)' % SEMIJOIN_TABLES[0]);
|
|
cls.client.execute('insert into %s values(2,4,30)' % SEMIJOIN_TABLES[0]);
|
|
cls.client.execute('insert into %s values(2,NULL,20)' % SEMIJOIN_TABLES[0]);
|
|
# loads some values with NULLs in the second table
|
|
cls.client.execute('insert into %s values(1,1,1)' % SEMIJOIN_TABLES[1]);
|
|
cls.client.execute('insert into %s values(1,1,10)' % SEMIJOIN_TABLES[1]);
|
|
cls.client.execute('insert into %s values(1,2,5)' % SEMIJOIN_TABLES[1]);
|
|
cls.client.execute('insert into %s values(1,NULL,10)' % SEMIJOIN_TABLES[1]);
|
|
cls.client.execute('insert into %s values(2,10,NULL)' % SEMIJOIN_TABLES[1]);
|
|
cls.client.execute('insert into %s values(3,NULL,NULL)' % SEMIJOIN_TABLES[1]);
|
|
cls.client.execute('insert into %s values(3,NULL,50)' % SEMIJOIN_TABLES[1]);
|
|
|
|
@classmethod
|
|
def __cleanup_semi_join_tables(cls):
|
|
cls.client.execute('drop table if exists functional.SemiJoinTblA')
|
|
cls.client.execute('drop table if exists functional.SemiJoinTblB')
|
|
|
|
@pytest.mark.execute_serially
|
|
def test_semi_joins(self, vector):
|
|
new_vector = copy(vector)
|
|
new_vector.get_value('exec_option')['batch_size'] = vector.get_value('batch_size')
|
|
self.run_test_case('QueryTest/semi-joins', new_vector)
|
|
|
|
def test_semi_joins_exhaustive(self, vector):
|
|
if self.exploration_strategy() != 'exhaustive': pytest.skip()
|
|
new_vector = copy(vector)
|
|
self.run_test_case('QueryTest/semi-joins-exhaustive', new_vector)
|