external
/
citus
mirror of https://github.com/citusdata/citus.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix/task execution (#3941) 

* Not set TaskExecution with adaptive executor

Adaptive executor is using a utility method from task tracker for
repartition joins, however adaptive executor doesn't need taskExecution.
It is only used by task tracker. This causes a problem when explain
analyze is used because what taskExecution is pointing to might be
random.

We solve this by not setting taskExecution from adaptive executor. So it
will stay NULL as set by CreateTask.

* use same memory context as task for taskExecution

Co-authored-by: Jelte Fennema <github-tech@jeltef.nl>
pull/3944/head^2
SaitTalhaNisanci 2020-06-24 12:10:00 +03:00 committed by GitHub
parent ac3c646ed5
commit f458d1fd1c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
5 changed files with 111 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
src/backend/distributed/executor/multi_task_tracker_executor.c
View File

@ -185,7 +185,9 @@ MultiTaskTrackerExecute(Job *job)
* We walk over the task tree, and create a task execution struct for each
* task. We then associate the task with its execution and get back a list.
*/
List *taskAndExecutionList = TaskAndExecutionList(jobTaskList);
bool createTaskExecution = true;
List *taskAndExecutionList = CreateTaskListForJobTree(jobTaskList,
createTaskExecution);
/*
* We now count the number of "top level" tasks in the query tree. Once they
@ -446,13 +448,14 @@ MultiTaskTrackerExecute(Job *job)
/*
* TaskAndExecutionList visits all tasks in the job tree, starting with the given
* job's task list. For each visited task, the function creates a task execution
* struct, associates the task execution with the task, and adds the task and its
* execution to a list. The function then returns the list.
* CreateTaskListForJobTree visits all tasks in the job tree (by following dependentTaskList),
* starting with the given job's task list.
* If createTaskExecution is set to true, for each visited task,
* the function creates a task execution struct associates the task execution with the task,
* and adds the task and its execution to a list. The function then returns the list.
*/
List *
TaskAndExecutionList(List *jobTaskList)
CreateTaskListForJobTree(List *jobTaskList, bool createTaskExecution)
{
List *taskAndExecutionList = NIL;
const int topLevelTaskHashSize = 32;
@ -470,9 +473,15 @@ TaskAndExecutionList(List *jobTaskList)
Task *task = (Task *) linitial(taskQueue);
taskQueue = list_delete_first(taskQueue);
if (createTaskExecution)
{
MemoryContext oldContext = MemoryContextSwitchTo(GetMemoryChunkContext(task));
/* create task execution and associate it with task */
TaskExecution *taskExecution = InitTaskExecution(task, EXEC_TASK_UNASSIGNED);
MemoryContextSwitchTo(oldContext);
task->taskExecution = taskExecution;
}
taskAndExecutionList = lappend(taskAndExecutionList, task);

3
src/backend/distributed/executor/repartition_join_execution.c
View File

@ -63,7 +63,8 @@ ExecuteDependentTasks(List *topLevelTasks, Job *topLevelJob)
{
EnsureNoModificationsHaveBeenDone();
List *allTasks = TaskAndExecutionList(topLevelTasks);
bool createTaskExecution = false;
List *allTasks = CreateTaskListForJobTree(topLevelTasks, createTaskExecution);
EnsureCompatibleLocalExecutionState(allTasks);

2
src/include/distributed/multi_task_tracker_executor.h
View File

@ -3,6 +3,6 @@
#ifndef MULTI_TASK_TRACKER_EXECUTOR_H
#define MULTI_TASK_TRACKER_EXECUTOR_H
extern List * TaskAndExecutionList(List *jobTaskList);
extern List * CreateTaskListForJobTree(List *jobTaskList, bool viaTaskTracker);
#endif /* MULTI_TASK_TRACKER_EXECUTOR_H */

64
src/test/regress/expected/chbenchmark_all_queries.out
View File

@ -864,6 +864,70 @@ LOG: join order: [ "stock" ][ reference join "item" ][ dual partition join "ord
abc | def
(1 row)
\set default_analyze_flags '(ANALYZE on, COSTS off, TIMING off, SUMMARY off)'
EXPLAIN :default_analyze_flags SELECT
su_name,
su_address
FROM
supplier,
nation
WHERE su_suppkey in
(SELECT
mod(s_i_id * s_w_id, 10000)
FROM
stock,
order_line
WHERE s_i_id IN
(SELECT i_id
FROM item
WHERE i_data LIKE 'co%')
AND ol_i_id = s_i_id
AND ol_delivery_d > '2008-05-23 12:00:00' -- was 2010, but our order is in 2008
GROUP BY s_i_id, s_w_id, s_quantity
HAVING 2*s_quantity > sum(ol_quantity))
AND su_nationkey = n_nationkey
AND n_name = 'Germany'
ORDER BY su_name;
LOG: join order: [ "stock" ][ reference join "item" ][ dual partition join "order_line" ]
QUERY PLAN
---------------------------------------------------------------------
Custom Scan (Citus Adaptive) (actual rows=1 loops=1)
-> Distributed Subplan XXX_1
Intermediate Data Size: 30 bytes
Result destination: Send to 2 nodes
-> Custom Scan (Citus Adaptive) (actual rows=3 loops=1)
Task Count: 4
Tuple data received from nodes: 12 bytes
Tasks Shown: None, not supported for re-partition queries
-> MapMergeJob
Map Task Count: 4
Merge Task Count: 4
-> MapMergeJob
Map Task Count: 4
Merge Task Count: 4
Task Count: 1
Tuple data received from nodes: 28 bytes
Tasks Shown: All
-> Task
Tuple data received from node: 28 bytes
Node: host=localhost port=xxxxx dbname=regression
-> Sort (actual rows=1 loops=1)
Sort Key: supplier.su_name
Sort Method: quicksort Memory: 25kB
-> Nested Loop Semi Join (actual rows=1 loops=1)
Join Filter: (supplier.su_suppkey = intermediate_result.mod)
Rows Removed by Join Filter: 1
-> Hash Join (actual rows=1 loops=1)
Hash Cond: (supplier.su_nationkey = nation.n_nationkey)
-> Seq Scan on supplier_1650035 supplier (actual rows=11 loops=1)
-> Hash (actual rows=1 loops=1)
Buckets: 1024 Batches: 1 Memory Usage: 9kB
-> Seq Scan on nation_1650034 nation (actual rows=1 loops=1)
Filter: (n_name = 'Germany'::bpchar)
Rows Removed by Filter: 3
-> Function Scan on read_intermediate_result intermediate_result (actual rows=2 loops=1)
(35 rows)
-- Query 21
-- DATA SET DOES NOT COVER THIS QUERY
SELECT

26
src/test/regress/sql/chbenchmark_all_queries.sql
View File

@ -673,6 +673,32 @@ WHERE su_suppkey in
AND n_name = 'Germany'
ORDER BY su_name;
\set default_analyze_flags '(ANALYZE on, COSTS off, TIMING off, SUMMARY off)'
EXPLAIN :default_analyze_flags SELECT
su_name,
su_address
FROM
supplier,
nation
WHERE su_suppkey in
(SELECT
mod(s_i_id * s_w_id, 10000)
FROM
stock,
order_line
WHERE s_i_id IN
(SELECT i_id
FROM item
WHERE i_data LIKE 'co%')
AND ol_i_id = s_i_id
AND ol_delivery_d > '2008-05-23 12:00:00' -- was 2010, but our order is in 2008
GROUP BY s_i_id, s_w_id, s_quantity
HAVING 2*s_quantity > sum(ol_quantity))
AND su_nationkey = n_nationkey
AND n_name = 'Germany'
ORDER BY su_name;
-- Query 21
-- DATA SET DOES NOT COVER THIS QUERY
SELECT