set max_threads = 16; set prefer_localhost_replica = 1; set optimize_aggregation_in_order = 0; set max_block_size = 65505; -- { echoOn } explain pipeline select * from (select * from numbers(1e8) group by number) group by number; explain pipeline select * from (select * from numbers_mt(1e8) group by number) group by number; explain pipeline select * from (select * from numbers_mt(1e8) group by number) order by number; explain pipeline select number from remote('127.0.0.{1,2,3}', system, numbers_mt) group by number settings distributed_aggregation_memory_efficient = 1; explain pipeline select number from remote('127.0.0.{1,2,3}', system, numbers_mt) group by number settings distributed_aggregation_memory_efficient = 0; -- { echoOff } DROP TABLE IF EXISTS proj_agg_02343; CREATE TABLE proj_agg_02343 ( k1 UInt32, k2 UInt32, k3 UInt32, value UInt32, PROJECTION aaaa ( SELECT k1, k2, k3, sum(value) GROUP BY k1, k2, k3 ) ) ENGINE = MergeTree ORDER BY tuple(); INSERT INTO proj_agg_02343 SELECT 1, number % 2, number % 4, number FROM numbers(100000); OPTIMIZE TABLE proj_agg_02343 FINAL; -- { echoOn } explain pipeline SELECT k1, k3, sum(value) v FROM remote('127.0.0.{1,2}', currentDatabase(), proj_agg_02343) GROUP BY k1, k3 SETTINGS distributed_aggregation_memory_efficient = 0; explain pipeline SELECT k1, k3, sum(value) v FROM remote('127.0.0.{1,2}', currentDatabase(), proj_agg_02343) GROUP BY k1, k3 SETTINGS distributed_aggregation_memory_efficient = 1; -- { echoOff } create table t(a UInt64) engine = MergeTree order by (a); system stop merges t; create table dist_t as t engine = Distributed(test_cluster_two_shards, currentDatabase(), t, a % 2); system stop merges dist_t; insert into dist_t select number from numbers_mt(10); insert into dist_t select number from numbers_mt(10); -- { echoOn } explain pipeline select a from remote('127.0.0.{1,2}', currentDatabase(), dist_t) group by a settings max_threads = 2, distributed_aggregation_memory_efficient = 1;