postgres/timescaledb
1
0
Fork 0
mirror of https://github.com/timescale/timescaledb.git synced 2025-05-17 19:13:16 +08:00
Code Issues Packages Projects Releases Wiki Activity

Support pushing partial aggregates to data nodes

Browse Source 
This change modifies the timescale_fdw to allow aggregates on a subset
of partitioning dimensions to be pushed down to data node.  The
deparsing code has also been modified to wrap the pushed aggregate in
a _timescaledb_internal.partialize_agg call, which will return the
proper array of values which postgres can combine into the finalized
value.
This commit is contained in:
Brian Rowe 2019-06-07 11:32:27 -04:00 committed by Erik Nordström
parent 112a7cf8d0
commit fd75dec223
12 changed files with 911 additions and 225 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
tsl/src/fdw/deparse.c
View File

@ -87,6 +87,7 @@
#include "timescaledb_fdw.h"
#include "extension_constants.h"
#include "plan_expand_hypertable.h"
#include "partialize_finalize.h"
/*
* Global context for foreign_expr_walker's search of an expression tree.
@ -755,10 +756,6 @@ foreign_expr_walker(Node *node, foreign_glob_cxt *glob_cxt, foreign_loc_cxt *out
if (!IS_UPPER_REL(glob_cxt->foreignrel))
return false;
/* Only non-split aggregates are pushable. */
if (agg->aggsplit != AGGSPLIT_SIMPLE)
return false;
/* As usual, it must be shippable. */
if (!is_shippable(agg->aggfnoid, ProcedureRelationId, fpinfo))
return false;
@ -2762,14 +2759,16 @@ deparseAggref(Aggref *node, deparse_expr_cxt *context)
{
StringInfo buf = context->buf;
bool use_variadic;
/* Only basic, non-split aggregation accepted. */
Assert(node->aggsplit == AGGSPLIT_SIMPLE);
bool partial_agg = node->aggsplit != AGGSPLIT_SIMPLE;
Assert(node->aggsplit == AGGSPLIT_SIMPLE || node->aggsplit == AGGSPLIT_INITIAL_SERIAL);
/* Check if need to print VARIADIC (cf. ruleutils.c) */
use_variadic = node->aggvariadic;
/* Find aggregate name from aggfnoid which is a pg_proc entry */
if (partial_agg)
appendStringInfoString(buf, INTERNAL_SCHEMA_NAME "." PARTIALIZE_FUNC_NAME "(");
appendFunctionName(node->aggfnoid, context);
appendStringInfoChar(buf, '(');
@ -2843,7 +2842,7 @@ deparseAggref(Aggref *node, deparse_expr_cxt *context)
deparseExpr((Expr *) node->aggfilter, context);
}
appendStringInfoChar(buf, ')');
appendStringInfoString(buf, partial_agg ? "))" : ")");
}
/*

10
tsl/src/fdw/server_scan_plan.c
View File

@ -17,6 +17,7 @@
#include <optimizer/restrictinfo.h>
#include <access/sysattr.h>
#include <utils/memutils.h>
#include <foreign/fdwapi.h>
#include <export.h>
#include <chunk_server.h>
@ -116,6 +117,15 @@ build_server_rel(PlannerInfo *root, Index relid, Oid serverid, RelOptInfo *paren
rel->lateral_relids = parent->lateral_relids;
rel->serverid = serverid;
/* We need to use the FDW interface to get called by the planner for
* partial aggs. For some reason, the standard upper_paths_hook is never
* called for upper rels of type UPPERREL_PARTIAL_GROUP_AGG, which is odd
* (see end of PostgreSQL planner.c:create_partial_grouping_paths). Until
* this gets fixed in the PostgreSQL planner, we're forced to set
* fdwroutine here although we will scan this rel with a ServerScan and
* not a ForeignScan. */
rel->fdwroutine = GetFdwRoutineByServerId(serverid);
return rel;
}

17
tsl/src/fdw/timescaledb_fdw.c
View File

@ -31,6 +31,7 @@
#include <utils/selfuncs.h>
#include <commands/defrem.h>
#include <commands/explain.h>
#include <commands/extension.h>
#include <rewrite/rewriteManip.h>
#include <miscadmin.h>
#include <libpq-fe.h>
@ -66,6 +67,7 @@
#include "remote/data_format.h"
#include "guc.h"
#include "server_scan_plan.h"
#include "extension_constants.h"
/* Default CPU cost to start up a foreign query. */
#define DEFAULT_FDW_STARTUP_COST 100.0
@ -1106,6 +1108,8 @@ get_foreign_paths(PlannerInfo *root, RelOptInfo *baserel, Oid foreigntableid)
TsFdwRelationInfo *fpinfo = fdw_relation_info_get(baserel);
Path *path;
Assert(fpinfo->type != TS_FDW_RELATION_INFO_HYPERTABLE_SERVER);
if (fpinfo->type == TS_FDW_RELATION_INFO_HYPERTABLE)
{
if (ts_guc_enable_per_server_queries)
@ -2919,9 +2923,6 @@ add_foreign_grouping_paths(PlannerInfo *root, RelOptInfo *input_rel, RelOptInfo
if (!parse->groupClause && !parse->groupingSets && !parse->hasAggs && !root->hasHavingQual)
return;
Assert(extra->patype == PARTITIONWISE_AGGREGATE_NONE ||
extra->patype == PARTITIONWISE_AGGREGATE_FULL);
/* save the input_rel as outerrel in fpinfo */
fpinfo->outerrel = input_rel;
@ -2985,7 +2986,8 @@ fdw_create_upper_paths(TsFdwRelationInfo *input_fpinfo, PlannerInfo *root, Upper
/* Ignore stages we don't support; and skip any duplicate calls (i.e.,
* output_rel->fdw_private has already been set by a previous call to this
* function). */
if (stage != UPPERREL_GROUP_AGG || output_rel->fdw_private)
if ((stage != UPPERREL_GROUP_AGG && stage != UPPERREL_PARTIAL_GROUP_AGG) ||
output_rel->fdw_private)
return;
input_fpinfo = fdw_relation_info_alloc(output_rel, input_fpinfo->type);
@ -3034,6 +3036,13 @@ get_foreign_upper_paths(PlannerInfo *root, UpperRelationKind stage, RelOptInfo *
if (fpinfo == NULL)
return;
/* We abuse the FDW API's GetForeignUpperPaths callback because, for some
* reason, the regular create_upper_paths_hook is never called for
* partially grouped rels, so we cannot use if for server rels. See end of
* PostgreSQL planner.c:create_partial_grouping_paths(). */
if (fpinfo->type == TS_FDW_RELATION_INFO_HYPERTABLE_SERVER)
return server_scan_create_upper_paths(root, stage, input_rel, output_rel, extra);
return fdw_create_upper_paths(fpinfo,
root,
stage,

8
tsl/src/partialize_finalize.h
View File

@ -12,8 +12,10 @@
#include "compat.h"
Datum tsl_finalize_agg_sfunc(PG_FUNCTION_ARGS);
Datum tsl_finalize_agg_ffunc(PG_FUNCTION_ARGS);
Datum tsl_partialize_agg(PG_FUNCTION_ARGS);
#define PARTIALIZE_FUNC_NAME "partialize_agg"
extern Datum tsl_finalize_agg_sfunc(PG_FUNCTION_ARGS);
extern Datum tsl_finalize_agg_ffunc(PG_FUNCTION_ARGS);
extern Datum tsl_partialize_agg(PG_FUNCTION_ARGS);
#endif

461
tsl/test/expected/dist_partial_agg.out Normal file
View File

@ -0,0 +1,461 @@
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- Need to be super user to create extension and add servers
\c :TEST_DBNAME :ROLE_SUPERUSER;
-- Need explicit password for non-super users to connect
ALTER ROLE :ROLE_DEFAULT_CLUSTER_USER CREATEDB PASSWORD 'pass';
GRANT USAGE ON FOREIGN DATA WRAPPER timescaledb_fdw TO :ROLE_DEFAULT_CLUSTER_USER;
SET ROLE :ROLE_DEFAULT_CLUSTER_USER;
-- Cleanup from other potential tests that created these databases
SET client_min_messages TO ERROR;
DROP DATABASE IF EXISTS server_1;
DROP DATABASE IF EXISTS server_2;
DROP DATABASE IF EXISTS server_3;
SET client_min_messages TO NOTICE;
CREATE DATABASE server_1;
CREATE DATABASE server_2;
CREATE DATABASE server_3;
\c server_1
SET client_min_messages TO ERROR;
CREATE EXTENSION timescaledb;
CREATE TYPE custom_type AS (high int, low int);
\c server_2
SET client_min_messages TO ERROR;
CREATE EXTENSION timescaledb;
CREATE TYPE custom_type AS (high int, low int);
\c server_3
SET client_min_messages TO ERROR;
CREATE EXTENSION timescaledb;
CREATE TYPE custom_type AS (high int, low int);
\c :TEST_DBNAME :ROLE_SUPERUSER;
SET ROLE :ROLE_DEFAULT_CLUSTER_USER;
\set TEST_TABLE 'conditions'
\ir 'include/aggregate_table_create.sql'
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- This file creates a table with a lot of different types to allow a range of aggregate functions.
-- This does not include the creation of a corresponding hypertable, as we may want to vary how that is done.
CREATE TYPE custom_type AS (high int, low int);
CREATE TABLE :TEST_TABLE (
timec TIMESTAMPTZ NOT NULL,
location TEXT NOT NULL,
region TEXT NOT NULL,
temperature DOUBLE PRECISION NULL,
humidity DOUBLE PRECISION NULL,
lowp double precision NULL,
highp double precision null,
allnull double precision null,
highlow custom_type null,
bit_int smallint,
good_life boolean
);
-- Add servers using the TimescaleDB server management API
SELECT * FROM add_server('server_1', database => 'server_1', password => 'pass', if_not_exists => true);
NOTICE: remote server database "server_1" already exists, skipping
server_name | host | port | database | username | server_username | created
-------------+-----------+-------+----------+----------------------+----------------------+---------
server_1 | localhost | 15432 | server_1 | default_cluster_user | default_cluster_user | t
(1 row)
SELECT * FROM add_server('server_2', database => 'server_2', password => 'pass', if_not_exists => true);
NOTICE: remote server database "server_2" already exists, skipping
server_name | host | port | database | username | server_username | created
-------------+-----------+-------+----------+----------------------+----------------------+---------
server_2 | localhost | 15432 | server_2 | default_cluster_user | default_cluster_user | t
(1 row)
SELECT * FROM add_server('server_3', database => 'server_3', password => 'pass', if_not_exists => true);
NOTICE: remote server database "server_3" already exists, skipping
server_name | host | port | database | username | server_username | created
-------------+-----------+-------+----------+----------------------+----------------------+---------
server_3 | localhost | 15432 | server_3 | default_cluster_user | default_cluster_user | t
(1 row)
SELECT table_name FROM create_distributed_hypertable( 'conditions', 'timec', 'location', 3, chunk_time_interval => INTERVAL '1 day');
table_name
------------
conditions
(1 row)
-- We need a lot of data and a lot of chunks to make the planner push down all of the aggregates
\ir 'include/aggregate_table_populate.sql'
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- This files assumes the existence of some table with definition as seen in the aggregate_table.sql file.
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'POR', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SFO', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SAC', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SEA', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'TAC', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'NYC', 'north-east', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'BOS', 'north-east', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'CHI', 'midwest', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'MIN', 'midwest', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'DET', 'midwest', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'LA', 'west', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SDG', 'west', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'PHX', 'west', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'DAL', 'south', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'AUS', 'south', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
SET enable_partitionwise_aggregate = ON;
-- Run an explain on the aggregate queries to make sure expected aggregates are being pushed down.
-- Grouping by the paritioning column should result in full aggregate pushdown where possible,
-- while using a non-partitioning column should result in a partial pushdown
\set PREFIX 'EXPLAIN (VERBOSE, COSTS OFF)'
\set GROUPING 'location'
\ir 'include/aggregate_queries.sql'
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- This files assumes the existence of some table with definition as seen in the aggregate_table.sql file.
-- All of these should be able to be pushed down if enabled
:PREFIX SELECT :GROUPING,
min(allnull) as min_allnull,
max(temperature) as max_temp,
sum(temperature)+sum(humidity) as agg_sum_expr,
avg(humidity),
ROUND( CAST(stddev(humidity) AS NUMERIC), 1),
bit_and(bit_int),
bit_or(bit_int),
bool_and(good_life),
every(temperature > 0),
bool_or(good_life),
count(*) as count_rows,
count(temperature) as count_temp,
count(allnull) as count_zero,
ROUND( CAST(corr(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(covar_pop(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(covar_samp(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_avgx(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_avgy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_count(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_intercept(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_r2(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_slope(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_sxx(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_sxy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_syy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(stddev(temperature) AS NUMERIC), 1) as stddev_temp,
ROUND( CAST(stddev_pop(temperature) AS NUMERIC), 1),
ROUND( CAST(stddev_samp(temperature) AS NUMERIC), 1),
ROUND( CAST(variance(temperature) AS NUMERIC), 1),
ROUND( CAST(var_pop(temperature) AS NUMERIC), 1),
ROUND( CAST(var_samp(temperature) AS NUMERIC), 1),
last(temperature, timec) as last_temp,
histogram(temperature, 0, 100, 1)
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: conditions.location, (min(conditions.allnull)), (max(conditions.temperature)), ((sum(conditions.temperature) + sum(conditions.humidity))), (avg(conditions.humidity)), (round((stddev(conditions.humidity))::numeric, 1)), (bit_and(conditions.bit_int)), (bit_or(conditions.bit_int)), (bool_and(conditions.good_life)), (every((conditions.temperature > '0'::double precision))), (bool_or(conditions.good_life)), (count(*)), (count(conditions.temperature)), (count(conditions.allnull)), (round((corr(conditions.temperature, conditions.humidity))::numeric, 1)), (round((covar_pop(conditions.temperature, conditions.humidity))::numeric, 1)), (round((covar_samp(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_avgx(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_avgy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_count(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_intercept(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_r2(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_slope(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_sxx(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_sxy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_syy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((stddev(conditions.temperature))::numeric, 1)), (round((stddev_pop(conditions.temperature))::numeric, 1)), (round((stddev_samp(conditions.temperature))::numeric, 1)), (round((variance(conditions.temperature))::numeric, 1)), (round((var_pop(conditions.temperature))::numeric, 1)), (round((var_samp(conditions.temperature))::numeric, 1)), (last(conditions.temperature, conditions.timec)), (histogram(conditions.temperature, '0'::double precision, '100'::double precision, 1))
Sort Key: conditions.location
-> Append
-> Custom Scan (ServerScan)
Output: conditions.location, (min(conditions.allnull)), (max(conditions.temperature)), ((sum(conditions.temperature) + sum(conditions.humidity))), (avg(conditions.humidity)), (round((stddev(conditions.humidity))::numeric, 1)), (bit_and(conditions.bit_int)), (bit_or(conditions.bit_int)), (bool_and(conditions.good_life)), (every((conditions.temperature > '0'::double precision))), (bool_or(conditions.good_life)), (count(*)), (count(conditions.temperature)), (count(conditions.allnull)), (round((corr(conditions.temperature, conditions.humidity))::numeric, 1)), (round((covar_pop(conditions.temperature, conditions.humidity))::numeric, 1)), (round((covar_samp(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_avgx(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_avgy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_count(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_intercept(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_r2(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_slope(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_sxx(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_sxy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_syy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((stddev(conditions.temperature))::numeric, 1)), (round((stddev_pop(conditions.temperature))::numeric, 1)), (round((stddev_samp(conditions.temperature))::numeric, 1)), (round((variance(conditions.temperature))::numeric, 1)), (round((var_pop(conditions.temperature))::numeric, 1)), (round((var_samp(conditions.temperature))::numeric, 1)), (last(conditions.temperature, conditions.timec)), (histogram(conditions.temperature, '0'::double precision, '100'::double precision, 1))
Relations: Aggregate on (public.conditions)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_2_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT location, min(allnull), max(temperature), (sum(temperature) + sum(humidity)), avg(humidity), round(stddev(humidity)::numeric, 1), bit_and(bit_int), bit_or(bit_int), bool_and(good_life), every((temperature > 0::double precision)), bool_or(good_life), count(*), count(temperature), count(allnull), round(corr(temperature, humidity)::numeric, 1), round(covar_pop(temperature, humidity)::numeric, 1), round(covar_samp(temperature, humidity)::numeric, 1), round(regr_avgx(temperature, humidity)::numeric, 1), round(regr_avgy(temperature, humidity)::numeric, 1), round(regr_count(temperature, humidity)::numeric, 1), round(regr_intercept(temperature, humidity)::numeric, 1), round(regr_r2(temperature, humidity)::numeric, 1), round(regr_slope(temperature, humidity)::numeric, 1), round(regr_sxx(temperature, humidity)::numeric, 1), round(regr_sxy(temperature, humidity)::numeric, 1), round(regr_syy(temperature, humidity)::numeric, 1), round(stddev(temperature)::numeric, 1), round(stddev_pop(temperature)::numeric, 1), round(stddev_samp(temperature)::numeric, 1), round(variance(temperature)::numeric, 1), round(var_pop(temperature)::numeric, 1), round(var_samp(temperature)::numeric, 1), public.last(temperature, timec), public.histogram(temperature, 0::double precision, 100::double precision, 1) FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: conditions_1.location, (min(conditions_1.allnull)), (max(conditions_1.temperature)), ((sum(conditions_1.temperature) + sum(conditions_1.humidity))), (avg(conditions_1.humidity)), (round((stddev(conditions_1.humidity))::numeric, 1)), (bit_and(conditions_1.bit_int)), (bit_or(conditions_1.bit_int)), (bool_and(conditions_1.good_life)), (every((conditions_1.temperature > '0'::double precision))), (bool_or(conditions_1.good_life)), (count(*)), (count(conditions_1.temperature)), (count(conditions_1.allnull)), (round((corr(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((covar_pop(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((covar_samp(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_avgx(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_avgy(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_count(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_intercept(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_r2(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_slope(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_sxx(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_sxy(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((regr_syy(conditions_1.temperature, conditions_1.humidity))::numeric, 1)), (round((stddev(conditions_1.temperature))::numeric, 1)), (round((stddev_pop(conditions_1.temperature))::numeric, 1)), (round((stddev_samp(conditions_1.temperature))::numeric, 1)), (round((variance(conditions_1.temperature))::numeric, 1)), (round((var_pop(conditions_1.temperature))::numeric, 1)), (round((var_samp(conditions_1.temperature))::numeric, 1)), (last(conditions_1.temperature, conditions_1.timec)), (histogram(conditions_1.temperature, '0'::double precision, '100'::double precision, 1))
Relations: Aggregate on (public.conditions)
Server: server_2
Chunks: _hyper_1_9_dist_chunk, _hyper_1_10_dist_chunk, _hyper_1_11_dist_chunk, _hyper_1_12_dist_chunk
Remote SQL: SELECT location, min(allnull), max(temperature), (sum(temperature) + sum(humidity)), avg(humidity), round(stddev(humidity)::numeric, 1), bit_and(bit_int), bit_or(bit_int), bool_and(good_life), every((temperature > 0::double precision)), bool_or(good_life), count(*), count(temperature), count(allnull), round(corr(temperature, humidity)::numeric, 1), round(covar_pop(temperature, humidity)::numeric, 1), round(covar_samp(temperature, humidity)::numeric, 1), round(regr_avgx(temperature, humidity)::numeric, 1), round(regr_avgy(temperature, humidity)::numeric, 1), round(regr_count(temperature, humidity)::numeric, 1), round(regr_intercept(temperature, humidity)::numeric, 1), round(regr_r2(temperature, humidity)::numeric, 1), round(regr_slope(temperature, humidity)::numeric, 1), round(regr_sxx(temperature, humidity)::numeric, 1), round(regr_sxy(temperature, humidity)::numeric, 1), round(regr_syy(temperature, humidity)::numeric, 1), round(stddev(temperature)::numeric, 1), round(stddev_pop(temperature)::numeric, 1), round(stddev_samp(temperature)::numeric, 1), round(variance(temperature)::numeric, 1), round(var_pop(temperature)::numeric, 1), round(var_samp(temperature)::numeric, 1), public.last(temperature, timec), public.histogram(temperature, 0::double precision, 100::double precision, 1) FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: conditions_2.location, (min(conditions_2.allnull)), (max(conditions_2.temperature)), ((sum(conditions_2.temperature) + sum(conditions_2.humidity))), (avg(conditions_2.humidity)), (round((stddev(conditions_2.humidity))::numeric, 1)), (bit_and(conditions_2.bit_int)), (bit_or(conditions_2.bit_int)), (bool_and(conditions_2.good_life)), (every((conditions_2.temperature > '0'::double precision))), (bool_or(conditions_2.good_life)), (count(*)), (count(conditions_2.temperature)), (count(conditions_2.allnull)), (round((corr(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((covar_pop(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((covar_samp(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_avgx(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_avgy(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_count(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_intercept(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_r2(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_slope(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_sxx(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_sxy(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((regr_syy(conditions_2.temperature, conditions_2.humidity))::numeric, 1)), (round((stddev(conditions_2.temperature))::numeric, 1)), (round((stddev_pop(conditions_2.temperature))::numeric, 1)), (round((stddev_samp(conditions_2.temperature))::numeric, 1)), (round((variance(conditions_2.temperature))::numeric, 1)), (round((var_pop(conditions_2.temperature))::numeric, 1)), (round((var_samp(conditions_2.temperature))::numeric, 1)), (last(conditions_2.temperature, conditions_2.timec)), (histogram(conditions_2.temperature, '0'::double precision, '100'::double precision, 1))
Relations: Aggregate on (public.conditions)
Server: server_3
Chunks: _hyper_1_5_dist_chunk, _hyper_1_6_dist_chunk, _hyper_1_7_dist_chunk, _hyper_1_8_dist_chunk
Remote SQL: SELECT location, min(allnull), max(temperature), (sum(temperature) + sum(humidity)), avg(humidity), round(stddev(humidity)::numeric, 1), bit_and(bit_int), bit_or(bit_int), bool_and(good_life), every((temperature > 0::double precision)), bool_or(good_life), count(*), count(temperature), count(allnull), round(corr(temperature, humidity)::numeric, 1), round(covar_pop(temperature, humidity)::numeric, 1), round(covar_samp(temperature, humidity)::numeric, 1), round(regr_avgx(temperature, humidity)::numeric, 1), round(regr_avgy(temperature, humidity)::numeric, 1), round(regr_count(temperature, humidity)::numeric, 1), round(regr_intercept(temperature, humidity)::numeric, 1), round(regr_r2(temperature, humidity)::numeric, 1), round(regr_slope(temperature, humidity)::numeric, 1), round(regr_sxx(temperature, humidity)::numeric, 1), round(regr_sxy(temperature, humidity)::numeric, 1), round(regr_syy(temperature, humidity)::numeric, 1), round(stddev(temperature)::numeric, 1), round(stddev_pop(temperature)::numeric, 1), round(stddev_samp(temperature)::numeric, 1), round(variance(temperature)::numeric, 1), round(var_pop(temperature)::numeric, 1), round(var_samp(temperature)::numeric, 1), public.last(temperature, timec), public.histogram(temperature, 0::double precision, 100::double precision, 1) FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) GROUP BY 1
(22 rows)
-- Aggregates on custom types are not yet pushed down
:PREFIX SELECT :GROUPING,
last(highlow, timec) as last_hl,
first(highlow, timec) as first_hl
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Merge Append
Sort Key: conditions.location
-> GroupAggregate
Output: conditions.location, last(conditions.highlow, conditions.timec), first(conditions.highlow, conditions.timec)
Group Key: conditions.location
-> Custom Scan (ServerScan) on public.conditions
Output: conditions.location, conditions.highlow, conditions.timec
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_2_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT timec, location, highlow FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) ORDER BY location ASC NULLS LAST
-> GroupAggregate
Output: conditions_1.location, last(conditions_1.highlow, conditions_1.timec), first(conditions_1.highlow, conditions_1.timec)
Group Key: conditions_1.location
-> Custom Scan (ServerScan) on public.conditions conditions_1
Output: conditions_1.location, conditions_1.highlow, conditions_1.timec
Server: server_2
Chunks: _hyper_1_9_dist_chunk, _hyper_1_10_dist_chunk, _hyper_1_11_dist_chunk, _hyper_1_12_dist_chunk
Remote SQL: SELECT timec, location, highlow FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) ORDER BY location ASC NULLS LAST
-> GroupAggregate
Output: conditions_2.location, last(conditions_2.highlow, conditions_2.timec), first(conditions_2.highlow, conditions_2.timec)
Group Key: conditions_2.location
-> Custom Scan (ServerScan) on public.conditions conditions_2
Output: conditions_2.location, conditions_2.highlow, conditions_2.timec
Server: server_3
Chunks: _hyper_1_5_dist_chunk, _hyper_1_6_dist_chunk, _hyper_1_7_dist_chunk, _hyper_1_8_dist_chunk
Remote SQL: SELECT timec, location, highlow FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) ORDER BY location ASC NULLS LAST
(26 rows)
-- Mix of aggregates that push down and those that don't
:PREFIX SELECT :GROUPING,
min(allnull) as min_allnull,
max(temperature) as max_temp,
sum(temperature)+sum(humidity) as agg_sum_expr,
avg(humidity),
ROUND( CAST(stddev(humidity) AS NUMERIC), 1),
bit_and(bit_int),
bit_or(bit_int),
bool_and(good_life),
every(temperature > 0),
bool_or(good_life),
first(highlow, timec) as first_hl
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Merge Append
Sort Key: conditions.location
-> GroupAggregate
Output: conditions.location, min(conditions.allnull), max(conditions.temperature), (sum(conditions.temperature) + sum(conditions.humidity)), avg(conditions.humidity), round((stddev(conditions.humidity))::numeric, 1), bit_and(conditions.bit_int), bit_or(conditions.bit_int), bool_and(conditions.good_life), every((conditions.temperature > '0'::double precision)), bool_or(conditions.good_life), first(conditions.highlow, conditions.timec)
Group Key: conditions.location
-> Custom Scan (ServerScan) on public.conditions
Output: conditions.location, conditions.allnull, conditions.temperature, conditions.humidity, conditions.bit_int, conditions.good_life, conditions.highlow, conditions.timec
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_2_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT timec, location, temperature, humidity, allnull, highlow, bit_int, good_life FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) ORDER BY location ASC NULLS LAST
-> GroupAggregate
Output: conditions_1.location, min(conditions_1.allnull), max(conditions_1.temperature), (sum(conditions_1.temperature) + sum(conditions_1.humidity)), avg(conditions_1.humidity), round((stddev(conditions_1.humidity))::numeric, 1), bit_and(conditions_1.bit_int), bit_or(conditions_1.bit_int), bool_and(conditions_1.good_life), every((conditions_1.temperature > '0'::double precision)), bool_or(conditions_1.good_life), first(conditions_1.highlow, conditions_1.timec)
Group Key: conditions_1.location
-> Custom Scan (ServerScan) on public.conditions conditions_1
Output: conditions_1.location, conditions_1.allnull, conditions_1.temperature, conditions_1.humidity, conditions_1.bit_int, conditions_1.good_life, conditions_1.highlow, conditions_1.timec
Server: server_2
Chunks: _hyper_1_9_dist_chunk, _hyper_1_10_dist_chunk, _hyper_1_11_dist_chunk, _hyper_1_12_dist_chunk
Remote SQL: SELECT timec, location, temperature, humidity, allnull, highlow, bit_int, good_life FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) ORDER BY location ASC NULLS LAST
-> GroupAggregate
Output: conditions_2.location, min(conditions_2.allnull), max(conditions_2.temperature), (sum(conditions_2.temperature) + sum(conditions_2.humidity)), avg(conditions_2.humidity), round((stddev(conditions_2.humidity))::numeric, 1), bit_and(conditions_2.bit_int), bit_or(conditions_2.bit_int), bool_and(conditions_2.good_life), every((conditions_2.temperature > '0'::double precision)), bool_or(conditions_2.good_life), first(conditions_2.highlow, conditions_2.timec)
Group Key: conditions_2.location
-> Custom Scan (ServerScan) on public.conditions conditions_2
Output: conditions_2.location, conditions_2.allnull, conditions_2.temperature, conditions_2.humidity, conditions_2.bit_int, conditions_2.good_life, conditions_2.highlow, conditions_2.timec
Server: server_3
Chunks: _hyper_1_5_dist_chunk, _hyper_1_6_dist_chunk, _hyper_1_7_dist_chunk, _hyper_1_8_dist_chunk
Remote SQL: SELECT timec, location, temperature, humidity, allnull, highlow, bit_int, good_life FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) ORDER BY location ASC NULLS LAST
(26 rows)
\set GROUPING 'region'
\ir 'include/aggregate_queries.sql'
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- This files assumes the existence of some table with definition as seen in the aggregate_table.sql file.
-- All of these should be able to be pushed down if enabled
:PREFIX SELECT :GROUPING,
min(allnull) as min_allnull,
max(temperature) as max_temp,
sum(temperature)+sum(humidity) as agg_sum_expr,
avg(humidity),
ROUND( CAST(stddev(humidity) AS NUMERIC), 1),
bit_and(bit_int),
bit_or(bit_int),
bool_and(good_life),
every(temperature > 0),
bool_or(good_life),
count(*) as count_rows,
count(temperature) as count_temp,
count(allnull) as count_zero,
ROUND( CAST(corr(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(covar_pop(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(covar_samp(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_avgx(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_avgy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_count(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_intercept(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_r2(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_slope(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_sxx(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_sxy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_syy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(stddev(temperature) AS NUMERIC), 1) as stddev_temp,
ROUND( CAST(stddev_pop(temperature) AS NUMERIC), 1),
ROUND( CAST(stddev_samp(temperature) AS NUMERIC), 1),
ROUND( CAST(variance(temperature) AS NUMERIC), 1),
ROUND( CAST(var_pop(temperature) AS NUMERIC), 1),
ROUND( CAST(var_samp(temperature) AS NUMERIC), 1),
last(temperature, timec) as last_temp,
histogram(temperature, 0, 100, 1)
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: conditions.region, (min(conditions.allnull)), (max(conditions.temperature)), ((sum(conditions.temperature) + sum(conditions.humidity))), (avg(conditions.humidity)), (round((stddev(conditions.humidity))::numeric, 1)), (bit_and(conditions.bit_int)), (bit_or(conditions.bit_int)), (bool_and(conditions.good_life)), (every((conditions.temperature > '0'::double precision))), (bool_or(conditions.good_life)), (count(*)), (count(conditions.temperature)), (count(conditions.allnull)), (round((corr(conditions.temperature, conditions.humidity))::numeric, 1)), (round((covar_pop(conditions.temperature, conditions.humidity))::numeric, 1)), (round((covar_samp(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_avgx(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_avgy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_count(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_intercept(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_r2(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_slope(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_sxx(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_sxy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((regr_syy(conditions.temperature, conditions.humidity))::numeric, 1)), (round((stddev(conditions.temperature))::numeric, 1)), (round((stddev_pop(conditions.temperature))::numeric, 1)), (round((stddev_samp(conditions.temperature))::numeric, 1)), (round((variance(conditions.temperature))::numeric, 1)), (round((var_pop(conditions.temperature))::numeric, 1)), (round((var_samp(conditions.temperature))::numeric, 1)), (last(conditions.temperature, conditions.timec)), (histogram(conditions.temperature, '0'::double precision, '100'::double precision, 1))
Sort Key: conditions.region
-> Finalize HashAggregate
Output: conditions.region, min(conditions.allnull), max(conditions.temperature), (sum(conditions.temperature) + sum(conditions.humidity)), avg(conditions.humidity), round((stddev(conditions.humidity))::numeric, 1), bit_and(conditions.bit_int), bit_or(conditions.bit_int), bool_and(conditions.good_life), every((conditions.temperature > '0'::double precision)), bool_or(conditions.good_life), count(*), count(conditions.temperature), count(conditions.allnull), round((corr(conditions.temperature, conditions.humidity))::numeric, 1), round((covar_pop(conditions.temperature, conditions.humidity))::numeric, 1), round((covar_samp(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_avgx(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_avgy(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_count(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_intercept(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_r2(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_slope(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_sxx(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_sxy(conditions.temperature, conditions.humidity))::numeric, 1), round((regr_syy(conditions.temperature, conditions.humidity))::numeric, 1), round((stddev(conditions.temperature))::numeric, 1), round((stddev_pop(conditions.temperature))::numeric, 1), round((stddev_samp(conditions.temperature))::numeric, 1), round((variance(conditions.temperature))::numeric, 1), round((var_pop(conditions.temperature))::numeric, 1), round((var_samp(conditions.temperature))::numeric, 1), last(conditions.temperature, conditions.timec), histogram(conditions.temperature, '0'::double precision, '100'::double precision, 1)
Group Key: conditions.region
-> Append
-> Custom Scan (ServerScan)
Output: conditions.region, (PARTIAL min(conditions.allnull)), (PARTIAL max(conditions.temperature)), (PARTIAL sum(conditions.temperature)), (PARTIAL sum(conditions.humidity)), (PARTIAL avg(conditions.humidity)), (PARTIAL stddev(conditions.humidity)), (PARTIAL bit_and(conditions.bit_int)), (PARTIAL bit_or(conditions.bit_int)), (PARTIAL bool_and(conditions.good_life)), (PARTIAL every((conditions.temperature > '0'::double precision))), (PARTIAL bool_or(conditions.good_life)), (PARTIAL count(*)), (PARTIAL count(conditions.temperature)), (PARTIAL count(conditions.allnull)), (PARTIAL corr(conditions.temperature, conditions.humidity)), (PARTIAL covar_pop(conditions.temperature, conditions.humidity)), (PARTIAL covar_samp(conditions.temperature, conditions.humidity)), (PARTIAL regr_avgx(conditions.temperature, conditions.humidity)), (PARTIAL regr_avgy(conditions.temperature, conditions.humidity)), (PARTIAL regr_count(conditions.temperature, conditions.humidity)), (PARTIAL regr_intercept(conditions.temperature, conditions.humidity)), (PARTIAL regr_r2(conditions.temperature, conditions.humidity)), (PARTIAL regr_slope(conditions.temperature, conditions.humidity)), (PARTIAL regr_sxx(conditions.temperature, conditions.humidity)), (PARTIAL regr_sxy(conditions.temperature, conditions.humidity)), (PARTIAL regr_syy(conditions.temperature, conditions.humidity)), (PARTIAL stddev(conditions.temperature)), (PARTIAL stddev_pop(conditions.temperature)), (PARTIAL stddev_samp(conditions.temperature)), (PARTIAL variance(conditions.temperature)), (PARTIAL var_pop(conditions.temperature)), (PARTIAL var_samp(conditions.temperature)), (PARTIAL last(conditions.temperature, conditions.timec)), (PARTIAL histogram(conditions.temperature, '0'::double precision, '100'::double precision, 1))
Relations: Aggregate on (public.conditions)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_2_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT region, _timescaledb_internal.partialize_agg(min(allnull)), _timescaledb_internal.partialize_agg(max(temperature)), _timescaledb_internal.partialize_agg(sum(temperature)), _timescaledb_internal.partialize_agg(sum(humidity)), _timescaledb_internal.partialize_agg(avg(humidity)), _timescaledb_internal.partialize_agg(stddev(humidity)), _timescaledb_internal.partialize_agg(bit_and(bit_int)), _timescaledb_internal.partialize_agg(bit_or(bit_int)), _timescaledb_internal.partialize_agg(bool_and(good_life)), _timescaledb_internal.partialize_agg(every((temperature > 0::double precision))), _timescaledb_internal.partialize_agg(bool_or(good_life)), _timescaledb_internal.partialize_agg(count(*)), _timescaledb_internal.partialize_agg(count(temperature)), _timescaledb_internal.partialize_agg(count(allnull)), _timescaledb_internal.partialize_agg(corr(temperature, humidity)), _timescaledb_internal.partialize_agg(covar_pop(temperature, humidity)), _timescaledb_internal.partialize_agg(covar_samp(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_avgx(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_avgy(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_count(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_intercept(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_r2(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_slope(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_sxx(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_sxy(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_syy(temperature, humidity)), _timescaledb_internal.partialize_agg(stddev(temperature)), _timescaledb_internal.partialize_agg(stddev_pop(temperature)), _timescaledb_internal.partialize_agg(stddev_samp(temperature)), _timescaledb_internal.partialize_agg(variance(temperature)), _timescaledb_internal.partialize_agg(var_pop(temperature)), _timescaledb_internal.partialize_agg(var_samp(temperature)), _timescaledb_internal.partialize_agg(public.last(temperature, timec)), _timescaledb_internal.partialize_agg(public.histogram(temperature, 0::double precision, 100::double precision, 1)) FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: conditions_1.region, (PARTIAL min(conditions_1.allnull)), (PARTIAL max(conditions_1.temperature)), (PARTIAL sum(conditions_1.temperature)), (PARTIAL sum(conditions_1.humidity)), (PARTIAL avg(conditions_1.humidity)), (PARTIAL stddev(conditions_1.humidity)), (PARTIAL bit_and(conditions_1.bit_int)), (PARTIAL bit_or(conditions_1.bit_int)), (PARTIAL bool_and(conditions_1.good_life)), (PARTIAL every((conditions_1.temperature > '0'::double precision))), (PARTIAL bool_or(conditions_1.good_life)), (PARTIAL count(*)), (PARTIAL count(conditions_1.temperature)), (PARTIAL count(conditions_1.allnull)), (PARTIAL corr(conditions_1.temperature, conditions_1.humidity)), (PARTIAL covar_pop(conditions_1.temperature, conditions_1.humidity)), (PARTIAL covar_samp(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_avgx(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_avgy(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_count(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_intercept(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_r2(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_slope(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_sxx(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_sxy(conditions_1.temperature, conditions_1.humidity)), (PARTIAL regr_syy(conditions_1.temperature, conditions_1.humidity)), (PARTIAL stddev(conditions_1.temperature)), (PARTIAL stddev_pop(conditions_1.temperature)), (PARTIAL stddev_samp(conditions_1.temperature)), (PARTIAL variance(conditions_1.temperature)), (PARTIAL var_pop(conditions_1.temperature)), (PARTIAL var_samp(conditions_1.temperature)), (PARTIAL last(conditions_1.temperature, conditions_1.timec)), (PARTIAL histogram(conditions_1.temperature, '0'::double precision, '100'::double precision, 1))
Relations: Aggregate on (public.conditions)
Server: server_2
Chunks: _hyper_1_9_dist_chunk, _hyper_1_10_dist_chunk, _hyper_1_11_dist_chunk, _hyper_1_12_dist_chunk
Remote SQL: SELECT region, _timescaledb_internal.partialize_agg(min(allnull)), _timescaledb_internal.partialize_agg(max(temperature)), _timescaledb_internal.partialize_agg(sum(temperature)), _timescaledb_internal.partialize_agg(sum(humidity)), _timescaledb_internal.partialize_agg(avg(humidity)), _timescaledb_internal.partialize_agg(stddev(humidity)), _timescaledb_internal.partialize_agg(bit_and(bit_int)), _timescaledb_internal.partialize_agg(bit_or(bit_int)), _timescaledb_internal.partialize_agg(bool_and(good_life)), _timescaledb_internal.partialize_agg(every((temperature > 0::double precision))), _timescaledb_internal.partialize_agg(bool_or(good_life)), _timescaledb_internal.partialize_agg(count(*)), _timescaledb_internal.partialize_agg(count(temperature)), _timescaledb_internal.partialize_agg(count(allnull)), _timescaledb_internal.partialize_agg(corr(temperature, humidity)), _timescaledb_internal.partialize_agg(covar_pop(temperature, humidity)), _timescaledb_internal.partialize_agg(covar_samp(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_avgx(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_avgy(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_count(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_intercept(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_r2(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_slope(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_sxx(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_sxy(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_syy(temperature, humidity)), _timescaledb_internal.partialize_agg(stddev(temperature)), _timescaledb_internal.partialize_agg(stddev_pop(temperature)), _timescaledb_internal.partialize_agg(stddev_samp(temperature)), _timescaledb_internal.partialize_agg(variance(temperature)), _timescaledb_internal.partialize_agg(var_pop(temperature)), _timescaledb_internal.partialize_agg(var_samp(temperature)), _timescaledb_internal.partialize_agg(public.last(temperature, timec)), _timescaledb_internal.partialize_agg(public.histogram(temperature, 0::double precision, 100::double precision, 1)) FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: conditions_2.region, (PARTIAL min(conditions_2.allnull)), (PARTIAL max(conditions_2.temperature)), (PARTIAL sum(conditions_2.temperature)), (PARTIAL sum(conditions_2.humidity)), (PARTIAL avg(conditions_2.humidity)), (PARTIAL stddev(conditions_2.humidity)), (PARTIAL bit_and(conditions_2.bit_int)), (PARTIAL bit_or(conditions_2.bit_int)), (PARTIAL bool_and(conditions_2.good_life)), (PARTIAL every((conditions_2.temperature > '0'::double precision))), (PARTIAL bool_or(conditions_2.good_life)), (PARTIAL count(*)), (PARTIAL count(conditions_2.temperature)), (PARTIAL count(conditions_2.allnull)), (PARTIAL corr(conditions_2.temperature, conditions_2.humidity)), (PARTIAL covar_pop(conditions_2.temperature, conditions_2.humidity)), (PARTIAL covar_samp(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_avgx(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_avgy(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_count(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_intercept(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_r2(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_slope(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_sxx(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_sxy(conditions_2.temperature, conditions_2.humidity)), (PARTIAL regr_syy(conditions_2.temperature, conditions_2.humidity)), (PARTIAL stddev(conditions_2.temperature)), (PARTIAL stddev_pop(conditions_2.temperature)), (PARTIAL stddev_samp(conditions_2.temperature)), (PARTIAL variance(conditions_2.temperature)), (PARTIAL var_pop(conditions_2.temperature)), (PARTIAL var_samp(conditions_2.temperature)), (PARTIAL last(conditions_2.temperature, conditions_2.timec)), (PARTIAL histogram(conditions_2.temperature, '0'::double precision, '100'::double precision, 1))
Relations: Aggregate on (public.conditions)
Server: server_3
Chunks: _hyper_1_5_dist_chunk, _hyper_1_6_dist_chunk, _hyper_1_7_dist_chunk, _hyper_1_8_dist_chunk
Remote SQL: SELECT region, _timescaledb_internal.partialize_agg(min(allnull)), _timescaledb_internal.partialize_agg(max(temperature)), _timescaledb_internal.partialize_agg(sum(temperature)), _timescaledb_internal.partialize_agg(sum(humidity)), _timescaledb_internal.partialize_agg(avg(humidity)), _timescaledb_internal.partialize_agg(stddev(humidity)), _timescaledb_internal.partialize_agg(bit_and(bit_int)), _timescaledb_internal.partialize_agg(bit_or(bit_int)), _timescaledb_internal.partialize_agg(bool_and(good_life)), _timescaledb_internal.partialize_agg(every((temperature > 0::double precision))), _timescaledb_internal.partialize_agg(bool_or(good_life)), _timescaledb_internal.partialize_agg(count(*)), _timescaledb_internal.partialize_agg(count(temperature)), _timescaledb_internal.partialize_agg(count(allnull)), _timescaledb_internal.partialize_agg(corr(temperature, humidity)), _timescaledb_internal.partialize_agg(covar_pop(temperature, humidity)), _timescaledb_internal.partialize_agg(covar_samp(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_avgx(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_avgy(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_count(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_intercept(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_r2(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_slope(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_sxx(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_sxy(temperature, humidity)), _timescaledb_internal.partialize_agg(regr_syy(temperature, humidity)), _timescaledb_internal.partialize_agg(stddev(temperature)), _timescaledb_internal.partialize_agg(stddev_pop(temperature)), _timescaledb_internal.partialize_agg(stddev_samp(temperature)), _timescaledb_internal.partialize_agg(variance(temperature)), _timescaledb_internal.partialize_agg(var_pop(temperature)), _timescaledb_internal.partialize_agg(var_samp(temperature)), _timescaledb_internal.partialize_agg(public.last(temperature, timec)), _timescaledb_internal.partialize_agg(public.histogram(temperature, 0::double precision, 100::double precision, 1)) FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4]) GROUP BY 1
(25 rows)
-- Aggregates on custom types are not yet pushed down
:PREFIX SELECT :GROUPING,
last(highlow, timec) as last_hl,
first(highlow, timec) as first_hl
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: conditions.region, (last(conditions.highlow, conditions.timec)), (first(conditions.highlow, conditions.timec))
Sort Key: conditions.region
-> Finalize HashAggregate
Output: conditions.region, last(conditions.highlow, conditions.timec), first(conditions.highlow, conditions.timec)
Group Key: conditions.region
-> Append
-> Partial HashAggregate
Output: conditions.region, PARTIAL last(conditions.highlow, conditions.timec), PARTIAL first(conditions.highlow, conditions.timec)
Group Key: conditions.region
-> Custom Scan (ServerScan) on public.conditions
Output: conditions.region, conditions.highlow, conditions.timec
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_2_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT timec, region, highlow FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4])
-> Partial HashAggregate
Output: conditions_1.region, PARTIAL last(conditions_1.highlow, conditions_1.timec), PARTIAL first(conditions_1.highlow, conditions_1.timec)
Group Key: conditions_1.region
-> Custom Scan (ServerScan) on public.conditions conditions_1
Output: conditions_1.region, conditions_1.highlow, conditions_1.timec
Server: server_2
Chunks: _hyper_1_9_dist_chunk, _hyper_1_10_dist_chunk, _hyper_1_11_dist_chunk, _hyper_1_12_dist_chunk
Remote SQL: SELECT timec, region, highlow FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4])
-> Partial HashAggregate
Output: conditions_2.region, PARTIAL last(conditions_2.highlow, conditions_2.timec), PARTIAL first(conditions_2.highlow, conditions_2.timec)
Group Key: conditions_2.region
-> Custom Scan (ServerScan) on public.conditions conditions_2
Output: conditions_2.region, conditions_2.highlow, conditions_2.timec
Server: server_3
Chunks: _hyper_1_5_dist_chunk, _hyper_1_6_dist_chunk, _hyper_1_7_dist_chunk, _hyper_1_8_dist_chunk
Remote SQL: SELECT timec, region, highlow FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4])
(31 rows)
-- Mix of aggregates that push down and those that don't
:PREFIX SELECT :GROUPING,
min(allnull) as min_allnull,
max(temperature) as max_temp,
sum(temperature)+sum(humidity) as agg_sum_expr,
avg(humidity),
ROUND( CAST(stddev(humidity) AS NUMERIC), 1),
bit_and(bit_int),
bit_or(bit_int),
bool_and(good_life),
every(temperature > 0),
bool_or(good_life),
first(highlow, timec) as first_hl
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: conditions.region, (min(conditions.allnull)), (max(conditions.temperature)), ((sum(conditions.temperature) + sum(conditions.humidity))), (avg(conditions.humidity)), (round((stddev(conditions.humidity))::numeric, 1)), (bit_and(conditions.bit_int)), (bit_or(conditions.bit_int)), (bool_and(conditions.good_life)), (every((conditions.temperature > '0'::double precision))), (bool_or(conditions.good_life)), (first(conditions.highlow, conditions.timec))
Sort Key: conditions.region
-> Finalize HashAggregate
Output: conditions.region, min(conditions.allnull), max(conditions.temperature), (sum(conditions.temperature) + sum(conditions.humidity)), avg(conditions.humidity), round((stddev(conditions.humidity))::numeric, 1), bit_and(conditions.bit_int), bit_or(conditions.bit_int), bool_and(conditions.good_life), every((conditions.temperature > '0'::double precision)), bool_or(conditions.good_life), first(conditions.highlow, conditions.timec)
Group Key: conditions.region
-> Append
-> Partial HashAggregate
Output: conditions.region, PARTIAL min(conditions.allnull), PARTIAL max(conditions.temperature), PARTIAL sum(conditions.temperature), PARTIAL sum(conditions.humidity), PARTIAL avg(conditions.humidity), PARTIAL stddev(conditions.humidity), PARTIAL bit_and(conditions.bit_int), PARTIAL bit_or(conditions.bit_int), PARTIAL bool_and(conditions.good_life), PARTIAL every((conditions.temperature > '0'::double precision)), PARTIAL bool_or(conditions.good_life), PARTIAL first(conditions.highlow, conditions.timec)
Group Key: conditions.region
-> Custom Scan (ServerScan) on public.conditions
Output: conditions.region, conditions.allnull, conditions.temperature, conditions.humidity, conditions.bit_int, conditions.good_life, conditions.highlow, conditions.timec
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_2_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT timec, region, temperature, humidity, allnull, highlow, bit_int, good_life FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4])
-> Partial HashAggregate
Output: conditions_1.region, PARTIAL min(conditions_1.allnull), PARTIAL max(conditions_1.temperature), PARTIAL sum(conditions_1.temperature), PARTIAL sum(conditions_1.humidity), PARTIAL avg(conditions_1.humidity), PARTIAL stddev(conditions_1.humidity), PARTIAL bit_and(conditions_1.bit_int), PARTIAL bit_or(conditions_1.bit_int), PARTIAL bool_and(conditions_1.good_life), PARTIAL every((conditions_1.temperature > '0'::double precision)), PARTIAL bool_or(conditions_1.good_life), PARTIAL first(conditions_1.highlow, conditions_1.timec)
Group Key: conditions_1.region
-> Custom Scan (ServerScan) on public.conditions conditions_1
Output: conditions_1.region, conditions_1.allnull, conditions_1.temperature, conditions_1.humidity, conditions_1.bit_int, conditions_1.good_life, conditions_1.highlow, conditions_1.timec
Server: server_2
Chunks: _hyper_1_9_dist_chunk, _hyper_1_10_dist_chunk, _hyper_1_11_dist_chunk, _hyper_1_12_dist_chunk
Remote SQL: SELECT timec, region, temperature, humidity, allnull, highlow, bit_int, good_life FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4])
-> Partial HashAggregate
Output: conditions_2.region, PARTIAL min(conditions_2.allnull), PARTIAL max(conditions_2.temperature), PARTIAL sum(conditions_2.temperature), PARTIAL sum(conditions_2.humidity), PARTIAL avg(conditions_2.humidity), PARTIAL stddev(conditions_2.humidity), PARTIAL bit_and(conditions_2.bit_int), PARTIAL bit_or(conditions_2.bit_int), PARTIAL bool_and(conditions_2.good_life), PARTIAL every((conditions_2.temperature > '0'::double precision)), PARTIAL bool_or(conditions_2.good_life), PARTIAL first(conditions_2.highlow, conditions_2.timec)
Group Key: conditions_2.region
-> Custom Scan (ServerScan) on public.conditions conditions_2
Output: conditions_2.region, conditions_2.allnull, conditions_2.temperature, conditions_2.humidity, conditions_2.bit_int, conditions_2.good_life, conditions_2.highlow, conditions_2.timec
Server: server_3
Chunks: _hyper_1_5_dist_chunk, _hyper_1_6_dist_chunk, _hyper_1_7_dist_chunk, _hyper_1_8_dist_chunk
Remote SQL: SELECT timec, region, temperature, humidity, allnull, highlow, bit_int, good_life FROM public.conditions WHERE _timescaledb_internal.chunks_in(conditions, ARRAY[1, 2, 3, 4])
(31 rows)
-- Full aggregate pushdown correctness check, compare location grouped query results with partionwise aggregates on and off
\set GROUPING 'location'
SELECT format('%s/results/dist_agg_loc_results_test.out', :'TEST_OUTPUT_DIR') as "RESULTS_TEST1",
format('%s/results/dist_agg_loc_results_control.out', :'TEST_OUTPUT_DIR') as "RESULTS_CONTROL1"
\gset
SELECT format('\! diff %s %s', :'RESULTS_CONTROL1', :'RESULTS_TEST1') as "DIFF_CMD1"
\gset
--generate the results into two different files
\set ECHO errors
:DIFF_CMD1
-- Partial aggregate pushdown correctness check, compare region grouped query results with partionwise aggregates on and off
\set GROUPING 'region'
SELECT format('%s/results/dist_agg_region_results_test.out', :'TEST_OUTPUT_DIR') as "RESULTS_TEST2",
format('%s/results/dist_agg_region_results_control.out', :'TEST_OUTPUT_DIR') as "RESULTS_CONTROL2"
\gset
SELECT format('\! diff %s %s', :'RESULTS_CONTROL2', :'RESULTS_TEST2') as "DIFF_CMD2"
\gset
--generate the results into two different files
\set ECHO errors
:DIFF_CMD2

140
tsl/test/expected/partitionwise_distributed-11.out
View File

@ -473,7 +473,7 @@ WHERE time < '2018-06-01 00:00'
GROUP BY 1
ORDER BY 1;
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: hyper."time", (avg(hyper.temp))
Sort Key: hyper."time"
@ -481,23 +481,19 @@ ORDER BY 1;
Output: hyper."time", avg(hyper.temp)
Group Key: hyper."time"
-> Append
-> Partial HashAggregate
Output: hyper."time", PARTIAL avg(hyper.temp)
Group Key: hyper."time"
-> Custom Scan (ServerScan) on public.hyper
Output: hyper."time", hyper.temp
-> Custom Scan (ServerScan)
Output: hyper."time", (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_3_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
-> Partial HashAggregate
Output: hyper_1."time", PARTIAL avg(hyper_1.temp)
Group Key: hyper_1."time"
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1."time", hyper_1.temp
Remote SQL: SELECT "time", _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1."time", (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_2_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
(23 rows)
Remote SQL: SELECT "time", _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
(19 rows)
-- Show result
SELECT time, avg(temp)
@ -699,28 +695,24 @@ FROM hyper
WHERE time < '2018-06-01 00:00'
GROUP BY 1;
QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Finalize HashAggregate
Output: hyper.location, avg(hyper.temp)
Group Key: hyper.location
-> Append
-> Partial HashAggregate
Output: hyper.location, PARTIAL avg(hyper.temp)
Group Key: hyper.location
-> Custom Scan (ServerScan) on public.hyper
Output: hyper.location, hyper.temp
-> Custom Scan (ServerScan)
Output: hyper.location, (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_3_dist_chunk
Remote SQL: SELECT location, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
-> Partial HashAggregate
Output: hyper_1.location, PARTIAL avg(hyper_1.temp)
Group Key: hyper_1.location
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1.location, hyper_1.temp
Remote SQL: SELECT location, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1.location, (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_4_dist_chunk, _hyper_1_2_dist_chunk
Remote SQL: SELECT location, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 1]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
(20 rows)
Remote SQL: SELECT location, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 1]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
(16 rows)
-- Expand query across repartition boundary. This makes it unsafe to
-- push down the FULL agg, so should expect a PARTIAL agg on
@ -804,7 +796,7 @@ FROM hyper
GROUP BY 1
ORDER BY 1;
QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: hyper.device, (avg(hyper.temp))
Sort Key: hyper.device
@ -812,23 +804,19 @@ ORDER BY 1;
Output: hyper.device, avg(hyper.temp)
Group Key: hyper.device
-> Append
-> Partial HashAggregate
Output: hyper.device, PARTIAL avg(hyper.temp)
Group Key: hyper.device
-> Custom Scan (ServerScan) on public.hyper
Output: hyper.device, hyper.temp
-> Custom Scan (ServerScan)
Output: hyper.device, (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_5_dist_chunk, _hyper_1_7_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3, 4])
-> Partial HashAggregate
Output: hyper_1.device, PARTIAL avg(hyper_1.temp)
Group Key: hyper_1.device
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1.device, hyper_1.temp
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3, 4]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1.device, (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_2_dist_chunk, _hyper_1_4_dist_chunk, _hyper_1_6_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3])
(23 rows)
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3]) GROUP BY 1
(19 rows)
-- Show result
SELECT device, avg(temp)
@ -898,7 +886,7 @@ WHERE time > '2018-04-19 00:01'
GROUP BY 1
ORDER BY 1;
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: hyper.device, (avg(hyper.temp))
Sort Key: hyper.device
@ -906,23 +894,19 @@ ORDER BY 1;
Output: hyper.device, avg(hyper.temp)
Group Key: hyper.device
-> Append
-> Partial HashAggregate
Output: hyper.device, PARTIAL avg(hyper.temp)
Group Key: hyper.device
-> Custom Scan (ServerScan) on public.hyper
Output: hyper.device, hyper.temp
-> Custom Scan (ServerScan)
Output: hyper.device, (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_3_dist_chunk, _hyper_1_5_dist_chunk, _hyper_1_7_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3, 4]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone))
-> Partial HashAggregate
Output: hyper_1.device, PARTIAL avg(hyper_1.temp)
Group Key: hyper_1.device
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1.device, hyper_1.temp
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3, 4]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1.device, (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_4_dist_chunk, _hyper_1_6_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone))
(23 rows)
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone)) GROUP BY 1
(19 rows)
SELECT device, avg(temp)
FROM hyper
@ -2184,28 +2168,24 @@ SELECT time_bucket('1 day', time), avg(temp)
FROM hyper1d
GROUP BY 1;
QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Finalize HashAggregate
Output: (time_bucket('@ 1 day'::interval, hyper1d."time")), avg(hyper1d.temp)
Group Key: (time_bucket('@ 1 day'::interval, hyper1d."time"))
-> Append
-> Partial HashAggregate
Output: (time_bucket('@ 1 day'::interval, hyper1d."time")), PARTIAL avg(hyper1d.temp)
Group Key: time_bucket('@ 1 day'::interval, hyper1d."time")
-> Custom Scan (ServerScan) on public.hyper1d
Output: time_bucket('@ 1 day'::interval, hyper1d."time"), hyper1d.temp
-> Custom Scan (ServerScan)
Output: (time_bucket('@ 1 day'::interval, hyper1d."time")), (PARTIAL avg(hyper1d.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_1
Chunks: _hyper_2_8_dist_chunk, _hyper_2_10_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5, 6])
-> Partial HashAggregate
Output: (time_bucket('@ 1 day'::interval, hyper1d_1."time")), PARTIAL avg(hyper1d_1.temp)
Group Key: time_bucket('@ 1 day'::interval, hyper1d_1."time")
-> Custom Scan (ServerScan) on public.hyper1d hyper1d_1
Output: time_bucket('@ 1 day'::interval, hyper1d_1."time"), hyper1d_1.temp
Remote SQL: SELECT public.time_bucket('@ 1 day'::interval, "time"), _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5, 6]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: (time_bucket('@ 1 day'::interval, hyper1d_1."time")), (PARTIAL avg(hyper1d_1.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_2
Chunks: _hyper_2_9_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4])
(20 rows)
Remote SQL: SELECT public.time_bucket('@ 1 day'::interval, "time"), _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4]) GROUP BY 1
(16 rows)
--- Only one chunk in query => safe to fully push down although not on
--- a partitioning dimension.
@ -2233,20 +2213,22 @@ FROM hyper1d
WHERE time < '2018-06-01 00:00'
GROUP BY 1;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
HashAggregate
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Finalize HashAggregate
Output: hyper1d.device, avg(hyper1d.temp)
Group Key: hyper1d.device
-> Append
-> Custom Scan (ServerScan) on public.hyper1d
Output: hyper1d.device, hyper1d.temp
-> Custom Scan (ServerScan)
Output: hyper1d.device, (PARTIAL avg(hyper1d.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_1
Chunks: _hyper_2_8_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
-> Custom Scan (ServerScan) on public.hyper1d hyper1d_1
Output: hyper1d_1.device, hyper1d_1.temp
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper1d_1.device, (PARTIAL avg(hyper1d_1.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_2
Chunks: _hyper_2_9_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
(14 rows)
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
(16 rows)

140
tsl/test/expected/partitionwise_distributed-12.out
View File

@ -473,7 +473,7 @@ WHERE time < '2018-06-01 00:00'
GROUP BY 1
ORDER BY 1;
QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: hyper."time", (avg(hyper.temp))
Sort Key: hyper."time"
@ -481,23 +481,19 @@ ORDER BY 1;
Output: hyper."time", avg(hyper.temp)
Group Key: hyper."time"
-> Append
-> Partial HashAggregate
Output: hyper."time", PARTIAL avg(hyper.temp)
Group Key: hyper."time"
-> Custom Scan (ServerScan) on public.hyper
Output: hyper."time", hyper.temp
-> Custom Scan (ServerScan)
Output: hyper."time", (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_3_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
-> Partial HashAggregate
Output: hyper_1."time", PARTIAL avg(hyper_1.temp)
Group Key: hyper_1."time"
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1."time", hyper_1.temp
Remote SQL: SELECT "time", _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1."time", (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_2_dist_chunk, _hyper_1_4_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
(23 rows)
Remote SQL: SELECT "time", _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
(19 rows)
-- Show result
SELECT time, avg(temp)
@ -699,28 +695,24 @@ FROM hyper
WHERE time < '2018-06-01 00:00'
GROUP BY 1;
QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Finalize HashAggregate
Output: hyper.location, avg(hyper.temp)
Group Key: hyper.location
-> Append
-> Partial HashAggregate
Output: hyper.location, PARTIAL avg(hyper.temp)
Group Key: hyper.location
-> Custom Scan (ServerScan) on public.hyper
Output: hyper.location, hyper.temp
-> Custom Scan (ServerScan)
Output: hyper.location, (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_3_dist_chunk
Remote SQL: SELECT location, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
-> Partial HashAggregate
Output: hyper_1.location, PARTIAL avg(hyper_1.temp)
Group Key: hyper_1.location
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1.location, hyper_1.temp
Remote SQL: SELECT location, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1.location, (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_4_dist_chunk, _hyper_1_2_dist_chunk
Remote SQL: SELECT location, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 1]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
(20 rows)
Remote SQL: SELECT location, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 1]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
(16 rows)
-- Expand query across repartition boundary. This makes it unsafe to
-- push down the FULL agg, so should expect a PARTIAL agg on
@ -804,7 +796,7 @@ FROM hyper
GROUP BY 1
ORDER BY 1;
QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: hyper.device, (avg(hyper.temp))
Sort Key: hyper.device
@ -812,23 +804,19 @@ ORDER BY 1;
Output: hyper.device, avg(hyper.temp)
Group Key: hyper.device
-> Append
-> Partial HashAggregate
Output: hyper.device, PARTIAL avg(hyper.temp)
Group Key: hyper.device
-> Custom Scan (ServerScan) on public.hyper
Output: hyper.device, hyper.temp
-> Custom Scan (ServerScan)
Output: hyper.device, (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_1_dist_chunk, _hyper_1_3_dist_chunk, _hyper_1_5_dist_chunk, _hyper_1_7_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3, 4])
-> Partial HashAggregate
Output: hyper_1.device, PARTIAL avg(hyper_1.temp)
Group Key: hyper_1.device
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1.device, hyper_1.temp
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3, 4]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1.device, (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_2_dist_chunk, _hyper_1_4_dist_chunk, _hyper_1_6_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3])
(23 rows)
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[1, 2, 3]) GROUP BY 1
(19 rows)
-- Show result
SELECT device, avg(temp)
@ -898,7 +886,7 @@ WHERE time > '2018-04-19 00:01'
GROUP BY 1
ORDER BY 1;
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sort
Output: hyper.device, (avg(hyper.temp))
Sort Key: hyper.device
@ -906,23 +894,19 @@ ORDER BY 1;
Output: hyper.device, avg(hyper.temp)
Group Key: hyper.device
-> Append
-> Partial HashAggregate
Output: hyper.device, PARTIAL avg(hyper.temp)
Group Key: hyper.device
-> Custom Scan (ServerScan) on public.hyper
Output: hyper.device, hyper.temp
-> Custom Scan (ServerScan)
Output: hyper.device, (PARTIAL avg(hyper.temp))
Relations: Aggregate on (public.hyper)
Server: server_1
Chunks: _hyper_1_3_dist_chunk, _hyper_1_5_dist_chunk, _hyper_1_7_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3, 4]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone))
-> Partial HashAggregate
Output: hyper_1.device, PARTIAL avg(hyper_1.temp)
Group Key: hyper_1.device
-> Custom Scan (ServerScan) on public.hyper hyper_1
Output: hyper_1.device, hyper_1.temp
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3, 4]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper_1.device, (PARTIAL avg(hyper_1.temp))
Relations: Aggregate on (public.hyper)
Server: server_2
Chunks: _hyper_1_4_dist_chunk, _hyper_1_6_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone))
(23 rows)
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper WHERE _timescaledb_internal.chunks_in(hyper, ARRAY[2, 3]) AND (("time" > '2018-04-19 00:01:00-07'::timestamp with time zone)) GROUP BY 1
(19 rows)
SELECT device, avg(temp)
FROM hyper
@ -2174,28 +2158,24 @@ SELECT time_bucket('1 day', time), avg(temp)
FROM hyper1d
GROUP BY 1;
QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Finalize HashAggregate
Output: (time_bucket('@ 1 day'::interval, hyper1d."time")), avg(hyper1d.temp)
Group Key: (time_bucket('@ 1 day'::interval, hyper1d."time"))
-> Append
-> Partial HashAggregate
Output: (time_bucket('@ 1 day'::interval, hyper1d."time")), PARTIAL avg(hyper1d.temp)
Group Key: time_bucket('@ 1 day'::interval, hyper1d."time")
-> Custom Scan (ServerScan) on public.hyper1d
Output: time_bucket('@ 1 day'::interval, hyper1d."time"), hyper1d.temp
-> Custom Scan (ServerScan)
Output: (time_bucket('@ 1 day'::interval, hyper1d."time")), (PARTIAL avg(hyper1d.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_1
Chunks: _hyper_2_8_dist_chunk, _hyper_2_10_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5, 6])
-> Partial HashAggregate
Output: (time_bucket('@ 1 day'::interval, hyper1d_1."time")), PARTIAL avg(hyper1d_1.temp)
Group Key: time_bucket('@ 1 day'::interval, hyper1d_1."time")
-> Custom Scan (ServerScan) on public.hyper1d hyper1d_1
Output: time_bucket('@ 1 day'::interval, hyper1d_1."time"), hyper1d_1.temp
Remote SQL: SELECT public.time_bucket('@ 1 day'::interval, "time"), _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5, 6]) GROUP BY 1
-> Custom Scan (ServerScan)
Output: (time_bucket('@ 1 day'::interval, hyper1d_1."time")), (PARTIAL avg(hyper1d_1.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_2
Chunks: _hyper_2_9_dist_chunk
Remote SQL: SELECT "time", temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4])
(20 rows)
Remote SQL: SELECT public.time_bucket('@ 1 day'::interval, "time"), _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4]) GROUP BY 1
(16 rows)
--- Only one chunk in query => safe to fully push down although not on
--- a partitioning dimension.
@ -2222,20 +2202,22 @@ FROM hyper1d
WHERE time < '2018-06-01 00:00'
GROUP BY 1;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
HashAggregate
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Finalize HashAggregate
Output: hyper1d.device, avg(hyper1d.temp)
Group Key: hyper1d.device
-> Append
-> Custom Scan (ServerScan) on public.hyper1d
Output: hyper1d.device, hyper1d.temp
-> Custom Scan (ServerScan)
Output: hyper1d.device, (PARTIAL avg(hyper1d.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_1
Chunks: _hyper_2_8_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
-> Custom Scan (ServerScan) on public.hyper1d hyper1d_1
Output: hyper1d_1.device, hyper1d_1.temp
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[5]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
-> Custom Scan (ServerScan)
Output: hyper1d_1.device, (PARTIAL avg(hyper1d_1.temp))
Relations: Aggregate on (public.hyper1d)
Server: server_2
Chunks: _hyper_2_9_dist_chunk
Remote SQL: SELECT device, temp FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone))
(14 rows)
Remote SQL: SELECT device, _timescaledb_internal.partialize_agg(avg(temp)) FROM public.hyper1d WHERE _timescaledb_internal.chunks_in(hyper1d, ARRAY[4]) AND (("time" < '2018-06-01 00:00:00-07'::timestamp with time zone)) GROUP BY 1
(16 rows)

1
tsl/test/sql/CMakeLists.txt
View File

@ -83,6 +83,7 @@ if (PG_VERSION_SUPPORTS_MULTINODE)
list(APPEND TEST_FILES
chunk_api.sql
timescaledb_fdw.sql
dist_partial_agg.sql
)
list(APPEND TEST_FILES_DEBUG
deparse.sql

114
tsl/test/sql/dist_partial_agg.sql Normal file
View File

@ -0,0 +1,114 @@
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- Need to be super user to create extension and add servers
\c :TEST_DBNAME :ROLE_SUPERUSER;
-- Need explicit password for non-super users to connect
ALTER ROLE :ROLE_DEFAULT_CLUSTER_USER CREATEDB PASSWORD 'pass';
GRANT USAGE ON FOREIGN DATA WRAPPER timescaledb_fdw TO :ROLE_DEFAULT_CLUSTER_USER;
SET ROLE :ROLE_DEFAULT_CLUSTER_USER;
-- Cleanup from other potential tests that created these databases
SET client_min_messages TO ERROR;
DROP DATABASE IF EXISTS server_1;
DROP DATABASE IF EXISTS server_2;
DROP DATABASE IF EXISTS server_3;
SET client_min_messages TO NOTICE;
CREATE DATABASE server_1;
CREATE DATABASE server_2;
CREATE DATABASE server_3;
\c server_1
SET client_min_messages TO ERROR;
CREATE EXTENSION timescaledb;
CREATE TYPE custom_type AS (high int, low int);
\c server_2
SET client_min_messages TO ERROR;
CREATE EXTENSION timescaledb;
CREATE TYPE custom_type AS (high int, low int);
\c server_3
SET client_min_messages TO ERROR;
CREATE EXTENSION timescaledb;
CREATE TYPE custom_type AS (high int, low int);
\c :TEST_DBNAME :ROLE_SUPERUSER;
SET ROLE :ROLE_DEFAULT_CLUSTER_USER;
\set TEST_TABLE 'conditions'
\ir 'include/aggregate_table_create.sql'
-- Add servers using the TimescaleDB server management API
SELECT * FROM add_server('server_1', database => 'server_1', password => 'pass', if_not_exists => true);
SELECT * FROM add_server('server_2', database => 'server_2', password => 'pass', if_not_exists => true);
SELECT * FROM add_server('server_3', database => 'server_3', password => 'pass', if_not_exists => true);
SELECT table_name FROM create_distributed_hypertable( 'conditions', 'timec', 'location', 3, chunk_time_interval => INTERVAL '1 day');
-- We need a lot of data and a lot of chunks to make the planner push down all of the aggregates
\ir 'include/aggregate_table_populate.sql'
SET enable_partitionwise_aggregate = ON;
-- Run an explain on the aggregate queries to make sure expected aggregates are being pushed down.
-- Grouping by the paritioning column should result in full aggregate pushdown where possible,
-- while using a non-partitioning column should result in a partial pushdown
\set PREFIX 'EXPLAIN (VERBOSE, COSTS OFF)'
\set GROUPING 'location'
\ir 'include/aggregate_queries.sql'
\set GROUPING 'region'
\ir 'include/aggregate_queries.sql'
-- Full aggregate pushdown correctness check, compare location grouped query results with partionwise aggregates on and off
\set GROUPING 'location'
SELECT format('%s/results/dist_agg_loc_results_test.out', :'TEST_OUTPUT_DIR') as "RESULTS_TEST1",
format('%s/results/dist_agg_loc_results_control.out', :'TEST_OUTPUT_DIR') as "RESULTS_CONTROL1"
\gset
SELECT format('\! diff %s %s', :'RESULTS_CONTROL1', :'RESULTS_TEST1') as "DIFF_CMD1"
\gset
--generate the results into two different files
\set ECHO errors
SET client_min_messages TO error;
--make output contain query results
\set PREFIX ''
\o :RESULTS_CONTROL1
SET enable_partitionwise_aggregate = OFF;
\ir 'include/aggregate_queries.sql'
\o
\o :RESULTS_TEST1
SET enable_partitionwise_aggregate = ON;
\ir 'include/aggregate_queries.sql'
\o
\set ECHO all
:DIFF_CMD1
-- Partial aggregate pushdown correctness check, compare region grouped query results with partionwise aggregates on and off
\set GROUPING 'region'
SELECT format('%s/results/dist_agg_region_results_test.out', :'TEST_OUTPUT_DIR') as "RESULTS_TEST2",
format('%s/results/dist_agg_region_results_control.out', :'TEST_OUTPUT_DIR') as "RESULTS_CONTROL2"
\gset
SELECT format('\! diff %s %s', :'RESULTS_CONTROL2', :'RESULTS_TEST2') as "DIFF_CMD2"
\gset
--generate the results into two different files
\set ECHO errors
SET client_min_messages TO error;
--make output contain query results
\set PREFIX ''
\o :RESULTS_CONTROL2
SET enable_partitionwise_aggregate = OFF;
\ir 'include/aggregate_queries.sql'
\o
\o :RESULTS_TEST2
SET enable_partitionwise_aggregate = ON;
\ir 'include/aggregate_queries.sql'
\o
\set ECHO all
:DIFF_CMD2

69
tsl/test/sql/include/aggregate_queries.sql Normal file
View File

@ -0,0 +1,69 @@
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- This files assumes the existence of some table with definition as seen in the aggregate_table.sql file.
-- All of these should be able to be pushed down if enabled
:PREFIX SELECT :GROUPING,
min(allnull) as min_allnull,
max(temperature) as max_temp,
sum(temperature)+sum(humidity) as agg_sum_expr,
avg(humidity),
ROUND( CAST(stddev(humidity) AS NUMERIC), 1),
bit_and(bit_int),
bit_or(bit_int),
bool_and(good_life),
every(temperature > 0),
bool_or(good_life),
count(*) as count_rows,
count(temperature) as count_temp,
count(allnull) as count_zero,
ROUND( CAST(corr(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(covar_pop(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(covar_samp(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_avgx(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_avgy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_count(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_intercept(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_r2(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_slope(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_sxx(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_sxy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(regr_syy(temperature, humidity) AS NUMERIC), 1),
ROUND( CAST(stddev(temperature) AS NUMERIC), 1) as stddev_temp,
ROUND( CAST(stddev_pop(temperature) AS NUMERIC), 1),
ROUND( CAST(stddev_samp(temperature) AS NUMERIC), 1),
ROUND( CAST(variance(temperature) AS NUMERIC), 1),
ROUND( CAST(var_pop(temperature) AS NUMERIC), 1),
ROUND( CAST(var_samp(temperature) AS NUMERIC), 1),
last(temperature, timec) as last_temp,
histogram(temperature, 0, 100, 1)
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
-- Aggregates on custom types are not yet pushed down
:PREFIX SELECT :GROUPING,
last(highlow, timec) as last_hl,
first(highlow, timec) as first_hl
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;
-- Mix of aggregates that push down and those that don't
:PREFIX SELECT :GROUPING,
min(allnull) as min_allnull,
max(temperature) as max_temp,
sum(temperature)+sum(humidity) as agg_sum_expr,
avg(humidity),
ROUND( CAST(stddev(humidity) AS NUMERIC), 1),
bit_and(bit_int),
bit_or(bit_int),
bool_and(good_life),
every(temperature > 0),
bool_or(good_life),
first(highlow, timec) as first_hl
FROM :TEST_TABLE
GROUP BY :GROUPING
ORDER BY :GROUPING;

22
tsl/test/sql/include/aggregate_table_create.sql Normal file
View File

@ -0,0 +1,22 @@
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- This file creates a table with a lot of different types to allow a range of aggregate functions.
-- This does not include the creation of a corresponding hypertable, as we may want to vary how that is done.
CREATE TYPE custom_type AS (high int, low int);
CREATE TABLE :TEST_TABLE (
timec TIMESTAMPTZ NOT NULL,
location TEXT NOT NULL,
region TEXT NOT NULL,
temperature DOUBLE PRECISION NULL,
humidity DOUBLE PRECISION NULL,
lowp double precision NULL,
highp double precision null,
allnull double precision null,
highlow custom_type null,
bit_int smallint,
good_life boolean
);

35
tsl/test/sql/include/aggregate_table_populate.sql Normal file
View File

@ -0,0 +1,35 @@
-- This file and its contents are licensed under the Timescale License.
-- Please see the included NOTICE for copyright information and
-- LICENSE-TIMESCALE for a copy of the license.
-- This files assumes the existence of some table with definition as seen in the aggregate_table.sql file.
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'POR', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SFO', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SAC', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SEA', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'TAC', 'west', generate_series(25, 85, 0.0625), 75, 40, 70, NULL, (1,2)::custom_type, 2, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'NYC', 'north-east', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'BOS', 'north-east', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'CHI', 'midwest', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'MIN', 'midwest', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'DET', 'midwest', generate_series(29, 41, 0.0125), 45, 50, 40, NULL, (3,4)::custom_type, 4, false;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'LA', 'west', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'SDG', 'west', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'PHX', 'west', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'DAL', 'south', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;
INSERT INTO :TEST_TABLE
SELECT generate_series('2018-12-01 00:00'::timestamp, '2018-12-04 08:00'::timestamp, '5 minute'), 'AUS', 'south', generate_series(61, 85, 0.025), 55, NULL, 28, NULL, NULL, 8, true;