def exec_and_compare_hive_and_impala_hs2(self, stmt, compare = lambda x, y: x == y):
"""Compare Hive and Impala results when executing the same statment over HS2"""
# execute_using_jdbc expects a Query object. Convert the query string into a Query
# object
query = Query()
query.query_str = stmt
# Run the statement targeting Hive
exec_opts = JdbcQueryExecConfig(impalad=HIVE_HS2_HOST_PORT, transport='SASL')
hive_results = execute_using_jdbc(query, exec_opts).data
# Run the statement targeting Impala
exec_opts = JdbcQueryExecConfig(impalad=IMPALAD_HS2_HOST_PORT, transport='NOSASL')
impala_results = execute_using_jdbc(query, exec_opts).data
# Compare the results
assert (impala_results is not None) and (hive_results is not None)
assert compare(impala_results, hive_results)
def exec_and_compare_hive_and_impala_hs2(self, stmt, compare = lambda x, y: x == y):
"""Compare Hive and Impala results when executing the same statment over HS2"""
# execute_using_jdbc expects a Query object. Convert the query string into a Query
# object
query = Query()
query.query_str = stmt
# Run the statement targeting Hive
exec_opts = JdbcQueryExecConfig(impalad=HIVE_HS2_HOST_PORT, transport='SASL')
hive_results = execute_using_jdbc(query, exec_opts).data
# Run the statement targeting Impala
exec_opts = JdbcQueryExecConfig(impalad=IMPALAD_HS2_HOST_PORT, transport='NOSASL')
impala_results = execute_using_jdbc(query, exec_opts).data
# Compare the results
assert (impala_results is not None) and (hive_results is not None)
assert compare(impala_results, hive_results)
def exec_and_compare_hive_and_impala_hs2(self, stmt):
"""Compare Hive and Impala results when executing the same statment over HS2"""
# execute_using_jdbc expects a Query object. Convert the query string into a Query
# object
query = Query()
query.query_str = stmt
# Run the statement targeting Hive
exec_opts = JdbcQueryExecConfig(impalad=HIVE_HS2_HOST_PORT)
hive_results = execute_using_jdbc(query, exec_opts).data
# Run the statement targeting Impala
exec_opts = JdbcQueryExecConfig(impalad=IMPALAD_HS2_HOST_PORT)
impala_results = execute_using_jdbc(query, exec_opts).data
# Compare the results
assert (impala_results is not None) and (hive_results is not None)
for impala, hive in zip(impala_results, hive_results):
assert impala == hive
def exec_and_compare_hive_and_impala_hs2(self, stmt):
"""Compare Hive and Impala results when executing the same statment over HS2"""
# execute_using_jdbc expects a Query object. Convert the query string into a Query
# object
query = Query()
query.query_str = stmt
# Run the statement targeting Hive
exec_opts = JdbcQueryExecConfig(impalad=HIVE_HS2_HOST_PORT)
hive_results = execute_using_jdbc(query, exec_opts).data
# Run the statement targeting Impala
exec_opts = JdbcQueryExecConfig(impalad=IMPALAD_HS2_HOST_PORT)
impala_results = execute_using_jdbc(query, exec_opts).data
# Compare the results
assert (impala_results is not None) and (hive_results is not None)
for impala, hive in zip(impala_results, hive_results):
assert impala == hive
def construct_queries(self, test_vector, scale_factor):
"""Transform a query map into a list of query objects.
Transform all the queries in the workload's query map to query objects based on the
input test vector and scale factor.
Args:
test_vector (?): query vector
scale_factor (str): eg. "300gb"
Returns:
(list of Query): these will be consumed by ?
"""
queries = list()
for query_name, query_str in self._query_map.iteritems():
queries.append(Query(name=query_name,
query_str=query_str,
workload=self._name,
scale_factor=scale_factor,
test_vector=test_vector))
return queries
