A copy will be executed locally if
- Local execution is enabled and current transaction accessed a local placement
- Local execution is enabled and we are inside a transaction block.
So even if local execution is enabled but we are not in a transaction block, the copy will not be run locally.
This will not run locally:
```
COPY distributed_table FROM STDIN;
....
```
This will run locally:
```
SET citus.enable_local_execution to 'on';
BEGIN;
COPY distributed_table FROM STDIN;
COMMIT;
....
```
.
There are 3 ways to do a copy in postgres programmatically:
- from a file
- from a program
- from a callback function
I have chosen to implement it with a callback function, which means that we write the rows of copy from a callback function to the output buffer, which is used to insert tuples into the actual table.
For each shard id, we have a buffer that keeps the current rows to be written, we perform the actual copy operation either when:
- copy buffer for the given shard id reaches to a threshold, which is currently 512KB
- we reach to the end of the copy
The buffer size is debatable(512KB). At a given time, we might allocate (local placement * buffer size) memory at most.
The local copy uses the same copy format as remote copy, which means that we serialize the data in the same format as remote copy and send it locally.
There was also the option to use ExecSimpleRelationInsert to insert
slots one by one, which would avoid the extra
serialization/deserialization but doing some benchmarks it seems that
using buffers are significantly better in terms of the performance.
You can see this comment for more details: https://github.com/citusdata/citus/pull/3557#discussion_r389499054
DESCRIPTION: Fix left join shard pruning in pushdown planner
Due to #2481 which moves outer join planning through the pushdown planner we caused a regression on the shard pruning behaviour for outer joins.
In the pushdown planner we make a union of the placement groups for all shards accessed by a query based on the filters we see during planning. Unfortunately implicit filters for left joins are not available during this part. This causes the inner part of an outer join to not prune any shards away. When we take the union of the placement groups it shows the behaviour of not having any shards pruned.
Since the inner part of an outer query will not return any rows if the outer part does not contain any rows we have observed we do not have to add the shard intervals of the inner part of an outer query to the list of shard intervals to query.
Fixes: #3512
Add failing tests, make changes to avoid crashes at least
Fix HAVING subquery pushdown ignoring reference table only subqueries,
also include HAVING in recursive planning
Given that we have a function IsDistributedTable which includes reference tables,
it seems best to have IsDistributedTableRTE & QueryContainsDistributedTableRTE
reflect that they do not include reference tables in their check
Similarly SublinkList's name should reflect that it only scans WHERE
contain_agg_clause asserts that we don't have SubLinks,
use contain_aggs_of_level as suggested by pg sourcecode
Before this commit, we considered !ContainsRecurringRTE() enough
for NotContainsOnlyRecurringTuples. However, instead, we can check
for existince of any distributed table.
DESCRIPTION: Fixes a bug that causes wrong results with complex outer joins
There are 2 problems with our early exit strategy that this commit fixes:
1- When we decide that a subplan results are sent to all worker nodes,
we used to skip traversing the whole distributed plan, instead of
skipping only the subplan.
2- We used to consider all available nodes in the cluster (secondaries
and inactive nodes as well as active primaries) when deciding on early
exit strategy. This resulted in failures to early exit when there are
secondaries or inactive nodes.
Semmle reported quite some places where we use a value that could be NULL. Most of these are not actually a real issue, but better to be on the safe side with these things and make the static analysis happy.
DESCRIPTION: Replace the query planner for the coordinator part with the postgres planner
Closes#2761
Citus had a simple rule based planner for the query executed on the query coordinator. This planner grew over time with the addigion of SQL support till it was getting close to the functionality of the postgres planner. Except the code was brittle and its complexity rose which made it hard to add new SQL support.
Given its resemblance with the postgres planner it was a long outstanding wish to replace our hand crafted planner with the well supported postgres planner. This patch replaces our planner with a call to postgres' planner.
Due to the functionality of the postgres planner we needed to support both projections and filters/quals on the citus custom scan node. When a sort operation is planned above the custom scan it might require fields to be reordered in the custom scan before returning the tuple (projection). The postgres planner assumes every custom scan node implements projections. Because we controlled the plan that was created we prevented reordering in the custom scan and never had implemented it before.
A same optimisation applies to having clauses that could have been where clauses. Instead of applying the filter as a having on the aggregate it will push it down into the plan which could reach a custom scan node.
For both filters and projections we have implemented them when tuples are read from the tuple store. If no projections or filters are required it will directly return the tuple from the tuple store. Otherwise it will loop tuples from the tuple store through the filter and projection until a tuple is found and returned.
Besides filters being pushed down a side effect of having quals that could have been a where clause is that a call to read intermediate result could be called before the first tuple is fetched from the custom scan. This failed because the intermediate result would only be pulled to the coordinator on the first tuple fetch. To overcome this problem we do run the distributed subplans now before we run the postgres executor. This ensures the intermediate result is present on the coordinator in time. We do account for total time instrumentation by removing the instrumentation before handing control to the psotgres executor and update the timings our self.
For future SQL support it is enough to create a valid query structure for the part of the query to be executed on the query coordinating node. As a utility we do serialise and print the query at debug level4 for engineers to inspect what kind of query is being planned on the query coordinator.
We don't actually use these functions anymore since merging #1477.
Advantages of removing:
1. They add work whenever we add a new node.
2. They contain some usage of stdlib APIs that are banned by Microsoft.
Removing it means we don't have to replace those with safe ones.
Comparison between differently sized integers in loop conditions can cause
infinite loops. This can happen when doing something like this:
```c
int64 very_big = MAX_INT32 + 1;
for (int32 i = 0; i < very_big; i++) {
// do something
}
// never reached because i overflows before it can reach the value of very_big
```
Previously a limitation in the shard pruning logic caused multi distribution value queries to always go into all the shards/workers whenever query also used OR conditions in WHERE clause.
Related to https://github.com/citusdata/citus/issues/2593 and https://github.com/citusdata/citus/issues/1537
There was no good workaround for this limitation. The limitation caused quite a bit of overhead with simple queries being sent to all workers/shards (especially with setups having lot of workers/shards).
An example of a previous plan which was inadequately pruned:
```
EXPLAIN SELECT count(*) FROM orders_hash_partitioned
WHERE (o_orderkey IN (1,2)) AND (o_custkey = 11 OR o_custkey = 22);
QUERY PLAN
---------------------------------------------------------------------
Aggregate (cost=0.00..0.00 rows=0 width=0)
-> Custom Scan (Citus Adaptive) (cost=0.00..0.00 rows=0 width=0)
Task Count: 4
Tasks Shown: One of 4
-> Task
Node: host=localhost port=xxxxx dbname=regression
-> Aggregate (cost=13.68..13.69 rows=1 width=8)
-> Seq Scan on orders_hash_partitioned_630000 orders_hash_partitioned (cost=0.00..13.68 rows=1 width=0)
Filter: ((o_orderkey = ANY ('{1,2}'::integer[])) AND ((o_custkey = 11) OR (o_custkey = 22)))
(9 rows)
```
After this commit the task count is what one would expect from the query defining multiple distinct values for the distribution column:
```
EXPLAIN SELECT count(*) FROM orders_hash_partitioned
WHERE (o_orderkey IN (1,2)) AND (o_custkey = 11 OR o_custkey = 22);
QUERY PLAN
---------------------------------------------------------------------
Aggregate (cost=0.00..0.00 rows=0 width=0)
-> Custom Scan (Citus Adaptive) (cost=0.00..0.00 rows=0 width=0)
Task Count: 2
Tasks Shown: One of 2
-> Task
Node: host=localhost port=xxxxx dbname=regression
-> Aggregate (cost=13.68..13.69 rows=1 width=8)
-> Seq Scan on orders_hash_partitioned_630000 orders_hash_partitioned (cost=0.00..13.68 rows=1 width=0)
Filter: ((o_orderkey = ANY ('{1,2}'::integer[])) AND ((o_custkey = 11) OR (o_custkey = 22)))
(9 rows)
```
"Core" of the pruning logic works as previously where it uses `PrunableInstances` to queue ORable valid constraints for shard pruning.
The difference is that now we build a compact internal representation of the query expression tree with PruningTreeNodes before actual shard pruning is run.
Pruning tree nodes represent boolean operators and the associated constraints of it. This internal format allows us to have compact representation of the query WHERE clauses which allows "core" pruning logic to work with OR-clauses correctly.
For example query having
`WHERE (o_orderkey IN (1,2)) AND (o_custkey=11 OR (o_shippriority > 1 AND o_shippriority < 10))`
gets transformed into:
1. AND(o_orderkey IN (1,2), OR(X, AND(X, X)))
2. AND(o_orderkey IN (1,2), OR(X, X))
3. AND(o_orderkey IN (1,2), X)
Here X is any set of unknown condition(s) for shard pruning.
This allow the final shard pruning to correctly recognize that shard pruning is done with the valid condition of `o_orderkey IN (1,2)`.
Another example with unprunable condition in query
`WHERE (o_orderkey IN (1,2)) OR (o_custkey=11 AND o_custkey=22)`
gets transformed into:
1. OR(o_orderkey IN (1,2), AND(X, X))
2. OR(o_orderkey IN (1,2), X)
Which is recognized as unprunable due to the OR condition between distribution column and unknown constraint -> goes to all shards.
Issue https://github.com/citusdata/citus/issues/1537 originally suggested transforming the query conditions into a full disjunctive normal form (DNF),
but this process of transforming into DNF is quite a heavy operation. It may "blow up" into a really large DNF form with complex queries having non trivial `WHERE` clauses.
I think the logic for shard pruning could be simplified further but I decided to leave the "core" of the shard pruning untouched.
The root of the problem is that, standard_planner() converts the following qual
```
{OPEXPR
:opno 98
:opfuncid 67
:opresulttype 16
:opretset false
:opcollid 0
:inputcollid 100
:args (
{VAR
:varno 1
:varattno 1
:vartype 25
:vartypmod -1
:varcollid 100
:varlevelsup 0
:varnoold 1
:varoattno 1
:location 45
}
{CONST
:consttype 25
:consttypmod -1
:constcollid 100
:constlen -1
:constbyval false
:constisnull true
:location 51
:constvalue <>
}
)
:location 49
}
```
To
```
(
{CONST
:consttype 16
:consttypmod -1
:constcollid 0
:constlen 1
:constbyval true
:constisnull true
:location -1
:constvalue <>
}
)
```
So, Citus doesn't deal with NULL values in real-time or non-fast path router queries.
And, in the FastPathRouter planner, we check constisnull in DistKeyInSimpleOpExpression().
However, in deferred pruning case, we do not check for isnull for const.
Thus, the fix consists of two parts:
- Let PruneShards() not crash when NULL parameter is passed
- For deferred shard pruning in fast-path queries, explicitly check that we have CONST which is not NULL
DESCRIPTION: Fix unnecessary repartition on joins with more than 4 tables
In 9.1 we have introduced support for all CH-benCHmark queries by widening our definitions of joins to include joins with expressions in them. This had the undesired side effect of Q5 regressing on its plan by implementing a repartition join.
It turned out this regression was not directly related to widening of the join clause, nor the schema employed by CH-benCHmark. Instead it had to do with 4 or more tables being joined in a chain. A chain meaning:
```sql
SELECT * FROM a,b,c,d WHERE a.part = b.part AND b.part = c.part AND ....
```
Due to how our join order planner was implemented it would only keep track of 1 of the partition columns when comparing if the join could be executed locally. This manifested in a join chain of 4 tables to _always_ be executed as a repartition join. 3 tables joined in a chain would have the middle table shared by the two outer tables causing the local join possibility to be found.
With this patch we keep a unique list (or set) of all partition columns participating in the join. When a candidate table is checked for a possibility to execute a local join it will check if there is any partition column in that set that matches an equality join clause on the partition column of the candidate table.
By taking into account all partition columns in the left relation it will now find the local join path on >= 4 tables joined in a chain.
fixes: #3276
Previously we only prevented AVG from being pushed down, but this is incorrect:
- array_agg, while somewhat non sensical to order by, will potentially be missing values
- combinefunc aggregation will raise errors about cstrings not being comparable (while we also can't know if the aggregate is commutative)
This commit limits approximating LIMIT pushdown when ordering by aggregates to:
min, max, sum, count, bit_and, bit_or, every, any
Which means of those we previously supported, we now exclude:
avg, array_agg, jsonb_agg, jsonb_object_agg, json_agg, json_object_agg, hll_add, hll_union, topn_add, topn_union
Previously, we've identified the usedSubPlans by only looking
to the subPlanId.
With this commit, we're expanding it to also include information
on the location of the subPlan.
This is useful to distinguish the cases where the subPlan is used
either on only HAVING or both HAVING and any other part of the query.
* Update shardPlacement->nodeId to uint
As the source of the shardPlacement->nodeId is always workerNode->nodeId,
and that is uint32.
We had this hack because of: 0ea4e52df5 (r266421409)
And, that is gone with: 90056f7d3c (diff-c532177d74c72d3f0e7cd10e448ab3c6L1123)
So, we're safe to do it now.
* Relax the restrictions on using the local execution
Previously, whenever any local execution happens, we disabled further
commands to do any remote queries. The basic motivation for doing that
is to prevent any accesses in the same transaction block to access the
same placements over multiple sessions: one is local session the other
is remote session to the same placement.
However, the current implementation does not distinguish local accesses
being to a placement or not. For example, we could have local accesses
that only touches intermediate results. In that case, we should not
implement the same restrictions as they become useless.
So, this is a pre-requisite for executing the intermediate result only
queries locally.
* Update the error messages
As the underlying implementation has changed, reflect it in the error
messages.
* Keep track of connections to local node
With this commit, we're adding infrastructure to track if any connection
to the same local host is done or not.
The main motivation for doing this is that we've previously were more
conservative about not choosing local execution. Simply, we disallowed
local execution if any connection to any remote node is done. However,
if we want to use local execution for intermediate result only queries,
this'd be annoying because we expect all queries to touch remote node
before the final query.
Note that this approach is still limiting in Citus MX case, but for now
we can ignore that.
* Formalize the concept of Local Node
Also some minor refactoring while creating the dummy placement
* Write intermediate results locally when the results are only needed locally
Before this commit, Citus used to always broadcast all the intermediate
results to remote nodes. However, it is possible to skip pushing
the results to remote nodes always.
There are two notable cases for doing that:
(a) When the query consists of only intermediate results
(b) When the query is a zero shard query
In both of the above cases, we don't need to access any data on the shards. So,
it is a valuable optimization to skip pushing the results to remote nodes.
The pattern mentioned in (a) is actually a common patterns that Citus users
use in practice. For example, if you have the following query:
WITH cte_1 AS (...), cte_2 AS (....), ... cte_n (...)
SELECT ... FROM cte_1 JOIN cte_2 .... JOIN cte_n ...;
The final query could be operating only on intermediate results. With this patch,
the intermediate results of the ctes are not unnecessarily pushed to remote
nodes.
* Add specific regression tests
As there are edge cases in Citus MX and with round-robin policy,
use the same queries on those cases as well.
* Fix failure tests
By forcing not to use local execution for intermediate results since
all the tests expects the results to be pushed remotely.
* Fix flaky test
* Apply code-review feedback
Mostly style changes
* Limit the max value of pg_dist_node_seq to reserve for internal use
Comment from code:
/*
* We had to implement this hack because on Postgres11 and below, the originalQuery
* and the query would have significant differences in terms of CTEs where CTEs
* would not be inlined on the query (as standard_planner() wouldn't inline CTEs
* on PG 11 and below).
*
* Instead, we prefer to pass the inlined query to the distributed planning. We rely
* on the fact that the query includes subqueries, and it'd definitely go through
* query pushdown planning. During query pushdown planning, the only relevant query
* tree is the original query.
*/
Deparsing and parsing a query can be heavy on CPU. When locally executing
the query we don't need to do this in theory most of the time.
This PR is the first step in allowing to skip deparsing and parsing
the query in these cases, by lazily creating the query string and
storing the query in the task. Future commits will make use of this and
not deparse and parse the query anymore, but use the one from the task
directly.
This is purely to enable better performance with prepared statements.
Before this commit, the fast path queries with prepared statements
where the distribution key includes a parameter always went through
distributed planning. After this change, we only go through distributed
planning on the first 5 executions.
In this commit, we're introducing a way to prevent CTE inlining via a GUC.
The GUC is used in all the tests where PG 11 and PG 12 tests would diverge
otherwise.
Note that, in PG 12, the restriction information for CTEs are generated. It
means that for some queries involving CTEs, Citus planner (router planner/
pushdown planner) may behave differently. So, via the GUC, we prevent
tests to diverge on PG 11 vs PG 12.
When we drop PG 11 support, we should get rid of the GUC, and mark
relevant ctes as MATERIALIZED, which does the same thing.
These set of tests has changed in both PG 11 and PG 12.
The changes are only about CTE inlining kicking in both
versions, and yielding the exact same distributed planning.
The idea is simple: Inline CTEs(if any), try distributed planning.
If the planning yields a successful distributed plan, simply return
it.
If the planning fails, fallback to distributed planning on the query
tree where CTEs are not inlined. In that case, if the planning failed
just because of the CTE inlining, via recursive planning, the same
query would yield a successful plan.
A very basic set of examples:
WITH cte_1 AS (SELECT * FROM test_table)
SELECT
*, row_number() OVER ()
FROM
cte_1;
or
WITH a AS (SELECT * FROM test_table),
b AS (SELECT * FROM test_table)
SELECT * FROM a JOIN b ON (a.value> b.value);
With this commit we add the necessary Citus function to inline CTEs
in a queryTree.
You might ask, why do we need to inline CTEs if Postgres is already
going to do it?
Few reasons behind this decision:
- One techinal node here is that Citus does the recursive CTE planning
by checking the originalQuery which is the query that has not gone
through the standard_planner().
CTEs in Citus is super powerful. It is practically key for full SQL
coverage for multi-shard queries. With CTEs, you can always reduce
any query multi-shard query into a router query via recursive
planning (thus full SQL coverage).
We cannot let CTE inlining break that. The main idea is Citus should
be able to retry planning if anything goes after CTE inlining.
So, by taking ownership of CTE inlining on the originalQuery, Citus
can fallback to recursive planning of CTEs if the planning with the
inlined query fails. It could have been a lot harder if we had relied
on standard_planner() to have the inlined CTEs on the original query.
- We want to have this feature in PostgreSQL 11 as well, but Postgres
only inlines in version 12
All the code in this commit is direct copy & paste from Postgres
source code.
We can classify the copy&paste code into two:
- Copy paste from CTE inline patch from postgres
(https://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=608b167f9f9c4553c35bb1ec0eab9ddae643989b)
These include the functions inline_cte(), inline_cte_walker(),
contain_dml(), contain_dml_walker().
It also include the code in function PostgreSQLCTEInlineCondition().
We prefer to extract that code into a seperate function, because
(a) we'll re-use the logic later (b) we added one check for PG_11
Finally, the struct "inline_cte_walker_context" is also copied from
the same Postgres commit.
- Copy paste from the other parts of the Postgres code
In order to implement CTE inlining in Postgres 12, the hackers
modified the query_tree_walker()/range_table_walker() with the
18c0da88a5
Since Citus needs to support the same logic in PG 11, we copy & pasted
that functions (and related flags) with the names pg_12_query_tree_walker()
and pg_12_range_table_walker()
* WIP
* wip
* add basic logic to run a single job with repartioning joins with adaptive executor
* fix some warnings and return in ExecuteDependedTasks if there is none
* Add the logic to run depended jobs in adaptive executor
The execution of depended tasks logic is changed. With the current
logic:
- All tasks are created from the top level task list.
- At one iteration:
- CurTasks whose dependencies are executed are found.
- CurTasks are executed in parallel with adapter executor main
logic.
- The iteration is repeated until all tasks are completed.
* Separate adaptive executor repartioning logic
* Remove duplicate parts
* cleanup directories and schemas
* add basic repartion tests for adaptive executor
* Use the first placement to fetch data
In task tracker, when there are replicas, we try to fetch from a replica
for which a map task is succeeded. TaskExecution is used for this,
however TaskExecution is not used in adaptive executor. So we cannot use
the same thing as task tracker.
Since adaptive executor fails when a map task fails (There is no retry
logic yet). We know that if we try to execute a fetch task, all of its
map tasks already succeeded, so we can just use the first one to fetch
from.
* fix clean directories logic
* do not change the search path while creating a udf
* Enable repartition joins with adaptive executor with only enable_reparitition_joins guc
* Add comments to adaptive_executor_repartition
* dont run adaptive executor repartition test in paralle with other tests
* execute cleanup only in the top level execution
* do cleanup only in the top level ezecution
* not begin a transaction if repartition query is used
* use new connections for repartititon specific queries
New connections are opened to send repartition specific queries. The
opened connections will be closed at the FinishDistributedExecution.
While sending repartition queries no transaction is begun so that
we can see all changes.
* error if a modification was done prior to repartition execution
* not start a transaction if a repartition query and sql task, and clean temporary files and schemas at each subplan level
* fix cleanup logic
* update tests
* add missing function comments
* add test for transaction with DDL before repartition query
* do not close repartition connections in adaptive executor
* rollback instead of commit in repartition join test
* use close connection instead of shutdown connection
* remove unnecesary connection list, ensure schema owner before removing directory
* rename ExecuteTaskListRepartition
* put fetch query string in planner not executor as we currently support only replication factor = 1 with adaptive executor and repartition query and we know the query string in the planner phase in that case
* split adaptive executor repartition to DAG execution logic and repartition logic
* apply review items
* apply review items
* use an enum for remote transaction state and fix cleanup for repartition
* add outside transaction flag to find connections that are unclaimed instead of always opening a new transaction
* fix style
* wip
* rename removejobdir to partition cleanup
* do not close connections at the end of repartition queries
* do repartition cleanup in pg catch
* apply review items
* decide whether to use transaction or not at execution creation
* rename isOutsideTransaction and add missing comment
* not error in pg catch while doing cleanup
* use replication factor of the creation time, not current time to decide if task tracker should be chosen
* apply review items
* apply review items
* apply review item
Use partition column's collation for range distributed tables
Don't allow non deterministic collations for hash distributed tables
CoPartitionedTables: don't compare unequal types
Previously,
- we'd push down ORDER BY, but this doesn't order intermediate results between workers
- we'd keep FILTER on master aggregate, which would raise an error about unexpected cstrings
Support for ARRAY[] expressions is limited to having a consistent shape,
eg ARRAY[(int,text),(int,text)] as opposed to ARRAY[(int,text),(float,text)] or ARRAY[(int,text),(int,text,float)]
Initialization of queryWindowClause and queryOrderByLimit "memset" underflow these variables.
It's possible due to the invalid usage sizeof this part of the program cause buffer overflow and function return data corruption in future changes.
In plain words, each distributed plan pulls the necessary intermediate
results to the worker nodes that the plan hits. This is primarily useful
in three ways.
(i) If the distributed plan that uses intermediate
result(s) is a router query, then the intermediate results are only
broadcasted to a single node.
(ii) If a distributed plan consists of only intermediate results, which
is not uncommon, the intermediate results are broadcasted to a single
node only.
(iii) If a distributed query hits a sub-set of the shards in multiple
workers, the intermediate results will be broadcasted to the relevant
node(s).
The final item (iii) becomes crucial for append/range distributed
tables where typically the distributed queries hit a small subset of
shards/workers.
To do this, for each query that Citus creates a distributed plan, we keep
track of the subPlans used in the queryTree, and save it in the distributed
plan. Just before Citus executes each subPlan, Citus first keeps track of
every worker node that the distributed plan hits, and marks every subPlan
should be broadcasted to these nodes. Later, for each subPlan which is a
distributed plan, Citus does this operation recursively since these
distributed plans may access to different subPlans, and those have to be
recorded as well.
DESCRIPTION: Expression in reference join
Fixed: #2582
This patch allows arbitrary expressions in the join clause when joining to a reference table. An example of such joins could be found in CHbenCHmark queries 7, 8, 9 and 11; `mod((s_w_id * s_i_id),10000) = su_suppkey` and `ascii(substr(c_state,1,1)) = n2.n_nationkey`. Since the join is on a reference table these queries are able to be pushed down to the workers.
To implement these queries we will widen the `IsJoinClause` predicate to not check if the expressions are a type `Var` after stripping the implicit coerciens. Instead we define a join clause when the `Var`'s in a clause come from more than 1 table.
This allows more clauses to pass into the logical planner's `MultiNodeTree(...)` planning function. To compensate for this we tighten down the `LocalJoin`, `SinglePartitionJoin` and `DualPartitionJoin` to check for direct column references when planning. This allows the planner to work with arbitrary join expressions on reference tables.
When the user picks "round-robin" policy, the aim is that the load
is distributed across nodes. However, for reference tables on the
coordinator, since local execution kicks in immediately, round-robin
is ignored.
With this change, we're excluding the placement on the coordinator.
Although the approach seems a little bit invasive because of
modifications in the placement list, that sounds acceptable.
We could have done this in some other ways such as:
1) Add a field to "Task->roundRobinPlacement" (or such), which is
updated as the first element after RoundRobinPolicy is applied.
During the execution, if that placement is local to the coordinator,
skip it and try the other remote placements.
2) On TaskAccessesLocalNode()@local_execution.c, check
task_assignment_policy, if round-robin selected and there is local
placement on the coordinator, skip it. However, task assignment is done
on planning, but this decision is happening on the execution, which
could create weird edge cases.
This change was actually already intended in #3124. However, the
postgres Makefile manually enables this warning too. This way we undo
that.
To confirm that it works two functions were changed to make use of not
having the warning anymore.
Phase 1 seeks to implement minimal infrastructure, so does not include:
- dynamic generation of support aggregates to handle multiple arguments
- configuration methods to direct aggregation strategy,
or mark an aggregate's serialize/deserialize as safe to operate across nodes
Aggregates can be distributed when:
- they have a single argument
- they have a combinefunc
- their transition type is not a pseudotype
This is necassery to support Q20 of the CHbenCHmark: #2582.
To summarize the fix: The subquery is converted into an INNER JOIN on a
table. This fixes the issue, since an INNER JOIN on a table is already
supported by the repartion planner.
The way this replacement is happening.:
1. Postgres replaces `col in (subquery)` with a SEMI JOIN (subquery) on col = subquery_result
2. If this subquery is simple enough Postgres will replace it with a
regular read from a table
3. If the subquery returns unique results (e.g. a primary key) Postgres
will convert the SEMI JOIN into an INNER JOIN during the planning. It
will not change this in the rewritten query though.
4. We check if Postgres sends us any SEMI JOINs during its join order
planning, if it doesn't we replace all SEMI JOINs in the rewritten
query with INNER JOIN (which we already support).
Postgres doesn't require you to add all columns that are in the target list to
the GROUP BY when you group by a unique column (or columns). It even actively
removes these group by clauses when you do.
This is normally fine, but for repartition joins it is not. The reason for this
is that the temporary tables don't have these primary key columns. So when the
worker executes the query it will complain that it is missing columns in the
group by.
This PR fixes that by adding an ANY_VALUE aggregate around each variable in
the target list that does is not contained in the group by or in an aggregate.
This is done only for repartition joins.
The ANY_VALUE aggregate chooses the value from an undefined row in the
group.
It looks like the logic to prevent RETURNING in reference tables to
have duplicate entries that comes from local and remote executions
leads to missing some tuples for distributed tables.
With this PR, we're ensuring to kick in the logic for reference tables
only.
* Remove unused executor codes
All of the codes of real-time executor. Some functions
in router executor still remains there because there
are common functions. We'll move them to accurate places
in the follow-up commits.
* Move GUCs to transaction mngnt and remove unused struct
* Update test output
* Get rid of references of real-time executor from code
* Warn if real-time executor is picked
* Remove lots of unused connection codes
* Removed unused code for connection restrictions
Real-time and router executors cannot handle re-using of the existing
connections within a transaction block.
Adaptive executor and COPY can re-use the connections. So, there is no
reason to keep the code around for applying the restrictions in the
placement connection logic.
We've changed the logic for pulling RTE_RELATIONs in #3109 and
non-colocated subquery joins and partitioned tables.
@onurctirtir found this steps where I traced back and found the issues.
While looking into it in more detail, we decided to expand the list in a
way that the callers get all the relevant RTE_RELATIONs RELKIND_RELATION,
RELKIND_PARTITIONED_TABLE, RELKIND_FOREIGN_TABLE and RELKIND_MATVIEW.
These are all relation kinds that Citus planner is aware of.
See #3125 for details on each item.
* Remove real-time/router executor tests-1
These are the ones which doesn't have '_%d' in the test
output files.
* Remove real-time/router executor tests-2
These are the ones which has in the test
output files.
* Move the tests outputs to correct place
* Make sure that single shard commits use 2PC on adaptive executor
It looks like we've messed the tests in #2891. Fixing back.
* Use adaptive executor for all router queries
This becomes important because when task-tracker is picked, we
used to pick router executor, which doesn't make sense.
* Remove explicit references to real-time/router executors in the tests
* JobExecutorType never picks real-time/router executors
* Make sure to go incremental in test output numbers
* Even users cannot pick real-time anymore
* Do not use real-time/router custom scans
* Get rid of unnecessary normalizations
* Reflect unneeded normalizations
* Get rid of unnecessary test output file
This completely hides `ListCell` to the user of the loop
Example usage:
```c
WorkerNode *workerNode = NULL;
foreach_ptr(workerNode, workerNodeList) {
// Do stuff with workerNode
}
```
Instead of:
```c
ListCell *workerNodeCell = NULL;
foreach(cell, workerNodeList) {
WorkerNode *workerNode = lfirst(workerNodeCell);
// Do stuff with workerNode
}
```
Areas for further optimization:
- Don't save subquery results to a local file on the coordinator when the subquery is not in the having clause
- Push the the HAVING with subquery to the workers if there's a group by on the distribution column
- Don't push down the results to the workers when we don't push down the HAVING clause, only the coordinator needs it
Fixes#520Fixes#756Closes#2047
When a function is marked as colocated with a distributed table,
we try delegating queries of kind "SELECT func(...)" to workers.
We currently only support this simple form, and don't delegate
forms like "SELECT f1(...), f2(...)", "SELECT f1(...) FROM ...",
or function calls inside transactions.
As a side effect, we also fix the transactional semantics of DO blocks.
Previously we didn't consider a DO block a multi-statement transaction.
Now we do.
Co-authored-by: Marco Slot <marco@citusdata.com>
Co-authored-by: serprex <serprex@users.noreply.github.com>
Co-authored-by: pykello <hadi.moshayedi@microsoft.com>
See a9c35cf85c
clang raises a warning due to FunctionCall2InfoData technically being variable sized
This is fine, as the struct is the size we want it to be. So silence the warning
This causes no behaviorial changes, only organizes better to implement modifying CTEs
Also rename ExtactInsertRangeTableEntry to ExtractResultRelationRTE,
as the source of this function didn't match the documentation
Remove Task's upsertQuery in favor of ROW_MODIFY_NONCOMMUTATIVE
Split up AcquireExecutorShardLock into more internal functions
Tests: Normalize multi_reference_table multi_create_table_constraints
With this commit, we're introducing the Adaptive Executor.
The commit message consists of two distinct sections. The first part explains
how the executor works. The second part consists of the commit messages of
the individual smaller commits that resulted in this commit. The readers
can search for the each of the smaller commit messages on
https://github.com/citusdata/citus and can learn more about the history
of the change.
/*-------------------------------------------------------------------------
*
* adaptive_executor.c
*
* The adaptive executor executes a list of tasks (queries on shards) over
* a connection pool per worker node. The results of the queries, if any,
* are written to a tuple store.
*
* The concepts in the executor are modelled in a set of structs:
*
* - DistributedExecution:
* Execution of a Task list over a set of WorkerPools.
* - WorkerPool
* Pool of WorkerSessions for the same worker which opportunistically
* executes "unassigned" tasks from a queue.
* - WorkerSession:
* Connection to a worker that is used to execute "assigned" tasks
* from a queue and may execute unasssigned tasks from the WorkerPool.
* - ShardCommandExecution:
* Execution of a Task across a list of placements.
* - TaskPlacementExecution:
* Execution of a Task on a specific placement.
* Used in the WorkerPool and WorkerSession queues.
*
* Every connection pool (WorkerPool) and every connection (WorkerSession)
* have a queue of tasks that are ready to execute (readyTaskQueue) and a
* queue/set of pending tasks that may become ready later in the execution
* (pendingTaskQueue). The tasks are wrapped in a ShardCommandExecution,
* which keeps track of the state of execution and is referenced from a
* TaskPlacementExecution, which is the data structure that is actually
* added to the queues and describes the state of the execution of a task
* on a particular worker node.
*
* When the task list is part of a bigger distributed transaction, the
* shards that are accessed or modified by the task may have already been
* accessed earlier in the transaction. We need to make sure we use the
* same connection since it may hold relevant locks or have uncommitted
* writes. In that case we "assign" the task to a connection by adding
* it to the task queue of specific connection (in
* AssignTasksToConnections). Otherwise we consider the task unassigned
* and add it to the task queue of a worker pool, which means that it
* can be executed over any connection in the pool.
*
* A task may be executed on multiple placements in case of a reference
* table or a replicated distributed table. Depending on the type of
* task, it may not be ready to be executed on a worker node immediately.
* For instance, INSERTs on a reference table are executed serially across
* placements to avoid deadlocks when concurrent INSERTs take conflicting
* locks. At the beginning, only the "first" placement is ready to execute
* and therefore added to the readyTaskQueue in the pool or connection.
* The remaining placements are added to the pendingTaskQueue. Once
* execution on the first placement is done the second placement moves
* from pendingTaskQueue to readyTaskQueue. The same approach is used to
* fail over read-only tasks to another placement.
*
* Once all the tasks are added to a queue, the main loop in
* RunDistributedExecution repeatedly does the following:
*
* For each pool:
* - ManageWorkPool evaluates whether to open additional connections
* based on the number unassigned tasks that are ready to execute
* and the targetPoolSize of the execution.
*
* Poll all connections:
* - We use a WaitEventSet that contains all (non-failed) connections
* and is rebuilt whenever the set of active connections or any of
* their wait flags change.
*
* We almost always check for WL_SOCKET_READABLE because a session
* can emit notices at any time during execution, but it will only
* wake up WaitEventSetWait when there are actual bytes to read.
*
* We check for WL_SOCKET_WRITEABLE just after sending bytes in case
* there is not enough space in the TCP buffer. Since a socket is
* almost always writable we also use WL_SOCKET_WRITEABLE as a
* mechanism to wake up WaitEventSetWait for non-I/O events, e.g.
* when a task moves from pending to ready.
*
* For each connection that is ready:
* - ConnectionStateMachine handles connection establishment and failure
* as well as command execution via TransactionStateMachine.
*
* When a connection is ready to execute a new task, it first checks its
* own readyTaskQueue and otherwise takes a task from the worker pool's
* readyTaskQueue (on a first-come-first-serve basis).
*
* In cases where the tasks finish quickly (e.g. <1ms), a single
* connection will often be sufficient to finish all tasks. It is
* therefore not necessary that all connections are established
* successfully or open a transaction (which may be blocked by an
* intermediate pgbouncer in transaction pooling mode). It is therefore
* essential that we take a task from the queue only after opening a
* transaction block.
*
* When a command on a worker finishes or the connection is lost, we call
* PlacementExecutionDone, which then updates the state of the task
* based on whether we need to run it on other placements. When a
* connection fails or all connections to a worker fail, we also call
* PlacementExecutionDone for all queued tasks to try the next placement
* and, if necessary, mark shard placements as inactive. If a task fails
* to execute on all placements, the execution fails and the distributed
* transaction rolls back.
*
* For multi-row INSERTs, tasks are executed sequentially by
* SequentialRunDistributedExecution instead of in parallel, which allows
* a high degree of concurrency without high risk of deadlocks.
* Conversely, multi-row UPDATE/DELETE/DDL commands take aggressive locks
* which forbids concurrency, but allows parallelism without high risk
* of deadlocks. Note that this is unrelated to SEQUENTIAL_CONNECTION,
* which indicates that we should use at most one connection per node, but
* can run tasks in parallel across nodes. This is used when there are
* writes to a reference table that has foreign keys from a distributed
* table.
*
* Execution finishes when all tasks are done, the query errors out, or
* the user cancels the query.
*
*-------------------------------------------------------------------------
*/
All the commits involved here:
* Initial unified executor prototype
* Latest changes
* Fix rebase conflicts to master branch
* Add missing variable for assertion
* Ensure that master_modify_multiple_shards() returns the affectedTupleCount
* Adjust intermediate result sizes
The real-time executor uses COPY command to get the results
from the worker nodes. Unified executor avoids that which
results in less data transfer. Simply adjust the tests to lower
sizes.
* Force one connection per placement (or co-located placements) when requested
The existing executors (real-time and router) always open 1 connection per
placement when parallel execution is requested.
That might be useful under certain circumstances:
(a) User wants to utilize as much as CPUs on the workers per
distributed query
(b) User has a transaction block which involves COPY command
Also, lots of regression tests rely on this execution semantics.
So, we'd enable few of the tests with this change as well.
* For parameters to be resolved before using them
For the details, see PostgreSQL's copyParamList()
* Unified executor sorts the returning output
* Ensure that unified executor doesn't ignore sequential execution of DDLJob's
Certain DDL commands, mainly creating foreign keys to reference tables,
should be executed sequentially. Otherwise, we'd end up with a self
distributed deadlock.
To overcome this situaiton, we set a flag `DDLJob->executeSequentially`
and execute it sequentially. Note that we have to do this because
the command might not be called within a transaction block, and
we cannot call `SetLocalMultiShardModifyModeToSequential()`.
This fixes at least two test: multi_insert_select_on_conflit.sql and
multi_foreign_key.sql
Also, I wouldn't mind scattering local `targetPoolSize` variables within
the code. The reason is that we'll soon have a GUC (or a global
variable based on a GUC) that'd set the pool size. In that case, we'd
simply replace `targetPoolSize` with the global variables.
* Fix 2PC conditions for DDL tasks
* Improve closing connections that are not fully established in unified execution
* Support foreign keys to reference tables in unified executor
The idea for supporting foreign keys to reference tables is simple:
Keep track of the relation accesses within a transaction block.
- If a parallel access happens on a distributed table which
has a foreign key to a reference table, one cannot modify
the reference table in the same transaction. Otherwise,
we're very likely to end-up with a self-distributed deadlock.
- If an access to a reference table happens, and then a parallel
access to a distributed table (which has a fkey to the reference
table) happens, we switch to sequential mode.
Unified executor misses the function calls that marks the relation
accesses during the execution. Thus, simply add the necessary calls
and let the logic kick in.
* Make sure to close the failed connections after the execution
* Improve comments
* Fix savepoints in unified executor.
* Rebuild the WaitEventSet only when necessary
* Unclaim connections on all errors.
* Improve failure handling for unified executor
- Implement the notion of errorOnAnyFailure. This is similar to
Critical Connections that the connection managament APIs provide
- If the nodes inside a modifying transaction expand, activate 2PC
- Fix few bugs related to wait event sets
- Mark placement INACTIVE during the execution as much as possible
as opposed to we do in the COMMIT handler
- Fix few bugs related to scheduling next placement executions
- Improve decision on when to use 2PC
Improve the logic to start a transaction block for distributed transactions
- Make sure that only reference table modifications are always
executed with distributed transactions
- Make sure that stored procedures and functions are executed
with distributed transactions
* Move waitEventSet to DistributedExecution
This could also be local to RunDistributedExecution(), but in that case
we had to mark it as "volatile" to avoid PG_TRY()/PG_CATCH() issues, and
cast it to non-volatile when doing WaitEventSetFree(). We thought that
would make code a bit harder to read than making this non-local, so we
move it here. See comments for PG_TRY() in postgres/src/include/elog.h
and "man 3 siglongjmp" for more context.
* Fix multi_insert_select test outputs
Two things:
1) One complex transaction block is now supported. Simply update
the test output
2) Due to dynamic nature of the unified executor, the orders of
the errors coming from the shards might change (e.g., all of
the queries on the shards would fail, but which one appears
on the error message?). To fix that, we simply added it to
our shardId normalization tool which happens just before diff.
* Fix subeury_and_cte test
The error message is updated from:
failed to execute task
To:
more than one row returned by a subquery or an expression
which is a lot clearer to the user.
* Fix intermediate_results test outputs
Simply update the error message from:
could not receive query results
to
result "squares" does not exist
which makes a lot more sense.
* Fix multi_function_in_join test
The error messages update from:
Failed to execute task XXX
To:
function f(..) does not exist
* Fix multi_query_directory_cleanup test
The unified executor does not create any intermediate files.
* Fix with_transactions test
A test case that just started to work fine
* Fix multi_router_planner test outputs
The error message is update from:
Could not receive query results
To:
Relation does not exists
which is a lot more clearer for the users
* Fix multi_router_planner_fast_path test
The error message is update from:
Could not receive query results
To:
Relation does not exists
which is a lot more clearer for the users
* Fix isolation_copy_placement_vs_modification by disabling select_opens_transaction_block
* Fix ordering in isolation_multi_shard_modify_vs_all
* Add executor locks to unified executor
* Make sure to allocate enought WaitEvents
The previous code was missing the waitEvents for the latch and
postmaster death.
* Fix rebase conflicts for master rebase
* Make sure that TRUNCATE relies on unified executor
* Implement true sequential execution for multi-row INSERTS
Execute the individual tasks executed one by one. Note that this is different than
MultiShardConnectionType == SEQUENTIAL_CONNECTION case (e.g., sequential execution
mode). In that case, running the tasks across the nodes in parallel is acceptable
and implemented in that way.
However, the executions that are qualified here would perform poorly if the
tasks across the workers are executed in parallel. We currently qualify only
one class of distributed queries here, multi-row INSERTs. If we do not enforce
true sequential execution, concurrent multi-row upserts could easily form
a distributed deadlock when the upserts touch the same rows.
* Remove SESSION_LIFESPAN flag in unified_executor
* Apply failure test updates
We've changed the failure behaviour a bit, and also the error messages
that show up to the user. This PR covers majority of the updates.
* Unified executor honors citus.node_connection_timeout
With this commit, unified executor errors out if even
a single connection cannot be established within
citus.node_connection_timeout.
And, as a side effect this fixes failure_connection_establishment
test.
* Properly increment/decrement pool size variables
Before this commit, the idle and active connection
counts were not properly calculated.
* insert_select_executor goes through unified executor.
* Add missing file for task tracker
* Modify ExecuteTaskListExtended()'s signature
* Sort output of INSERT ... SELECT ... RETURNING
* Take partition locks correctly in unified executor
* Alternative implementation for force_max_query_parallelization
* Fix compile warnings in unified executor
* Fix style issues
* Decrement idleConnectionCount when idle connection is lost
* Always rebuild the wait event sets
In the previous implementation, on waitFlag changes, we were only
modifying the wait events. However, we've realized that it might
be an over optimization since (a) we couldn't see any performance
benefits (b) we see some errors on failures and because of (a)
we prefer to disable it now.
* Make sure to allocate enough sized waitEventSet
With multi-row INSERTs, we might have more sessions than
task*workerCount after few calls of RunDistributedExecution()
because the previous sessions would also be alive.
Instead, re-allocate events when the connectino set changes.
* Implement SELECT FOR UPDATE on reference tables
On master branch, we do two extra things on SELECT FOR UPDATE
queries on reference tables:
- Acquire executor locks
- Execute the query on all replicas
With this commit, we're implementing the same logic on the
new executor.
* SELECT FOR UPDATE opens transaction block even if SelectOpensTransactionBlock disabled
Otherwise, users would be very confused and their logic is very likely
to break.
* Fix build error
* Fix the newConnectionCount calculation in ManageWorkerPool
* Fix rebase conflicts
* Fix minor test output differences
* Fix citus indent
* Remove duplicate sorts that is added with rebase
* Create distributed table via executor
* Fix wait flags in CheckConnectionReady
* failure_savepoints output for unified executor.
* failure_vacuum output (pg 10) for unified executor.
* Fix WaitEventSetWait timeout in unified executor
* Stabilize failure_truncate test output
* Add an ORDER BY to multi_upsert
* Fix regression test outputs after rebase to master
* Add executor.c comment
* Rename executor.c to adaptive_executor.c
* Do not schedule tasks if the failed placement is not ready to execute
Before the commit, we were blindly scheduling the next placement executions
even if the failed placement is not on the ready queue. Now, we're ensuring
that if failed placement execution is on a failed pool or session where the
execution is on the pendingQueue, we do not schedule the next task. Because
the other placement execution should be already running.
* Implement a proper custom scan node for adaptive executor
- Switch between the executors, add GUC to set the pool size
- Add non-adaptive regression test suites
- Enable CIRCLE CI for non-adaptive tests
- Adjust test output files
* Add slow start interval to the executor
* Expose max_cached_connection_per_worker to user
* Do not start slow when there are cached connections
* Consider ExecutorSlowStartInterval in NextEventTimeout
* Fix memory issues with ReceiveResults().
* Disable executor via TaskExecutorType
* Make sure to execute the tests with the other executor
* Use task_executor_type to enable-disable adaptive executor
* Remove useless code
* Adjust the regression tests
* Add slow start regression test
* Rebase to master
* Fix test failures in adaptive executor.
* Rebase to master - 2
* Improve comments & debug messages
* Set force_max_query_parallelization in isolation_citus_dist_activity
* Force max parallelization for creating shards when asked to use exclusive connection.
* Adjust the default pool size
* Expand description of max_adaptive_executor_pool_size GUC
* Update warnings in FinishRemoteTransactionCommit()
* Improve session clean up at the end of execution
Explicitly list all the states that the execution might end,
otherwise warn.
* Remove MULTI_CONNECTION_WAIT_RETRY which is not used at all
* Add more ORDER BYs to multi_mx_partitioning
If a query is router executable, it hits a single shard and therefore has a
single task associated with it. Therefore there is no need to sort the task list
that has a single element.
Also we already have a list of active shard placements, sending it in param
and reuse it.
We used to rely on PG function flatten_join_alias_vars
to resolve actual columns referenced in target entry list.
The function goes deep and finds the actual relation. This logic
usually works fine. However, when joins are given an alias, inner
relation names are not visible to target entry entry. Thus relation
resolving should stop when we the target entry column refers an
rte of an aliased join.
We stopped using PG function and provided our own flatten function.
Our assumption that strip_implicit_coercions would leave us with a bi-
nary-compatible type to that of the partition key was wrong. Instead,
we should ensure the RHS of the comparison we perform is proactively
coerced into a compatible type (at least binary compatible).
Do it in two ways (a) re-use the rte list as much as possible instead of
re-calculating over and over again (b) Limit the recursion to the relevant
parts of the query tree
Before this commit, shardPlacements were identified with shardId, nodeName
and nodeport. Instead of using nodeName and nodePort, we now use nodeId
since it apparently has performance benefits in several places in the
code.
The rule for infinite recursion is the following:
- If the query contains a subquery which is recursively planned, and
no other subqueries can be recursively planned due to correlation
(e.g., LATERAL joins), the planner keeps recursing again and again.
One interesting thing here is that even if a subquery contains only intermediate
result(s), we re-recursively plan that. In the end, the logic in the code does the following:
- Try recursive planning any of the subqueries in the query tree
- If any subquery is recursively planned, call the planner again
where the subquery is replaced with the intermediate result.
- Try recursively planning any of the queries
- If any subquery is recursively planned, call the planner again
where the subquery (in this case it is already intermediate result)
is replaced with the intermediate result.
- Try recursively planning any of the queries
- If any subquery is recursively planned, call the planner again
where the subquery (in this case it is already intermediate result)
is replaced with the intermediate result.
- Try recursively planning any of the queries
- If any subquery is recursively planned, call the planner again
where the subquery (in this case it is already intermediate result)
is replaced with the intermediate result.
......
Since flattening query may flatten outer joins' columns into coalesce expr that is
in the USING part, and that was not expected before this commit, these queries were
erroring out. It is fixed by this commit with considering coalesce expression as well.
Before this commit, round-robin task assignment policy was relying
on the taskId. Thus, even inside a transaction, the tasks were
assigned to different nodes. This was especially problematic
while reading from reference tables within transaction blocks.
Because, we had to expand the distributed transaction to many
nodes that are not necessarily already in the distributed transaction.
In this context, we define "Fast Path Planning for SELECT" as trivial
queries where Citus can skip relying on the standard_planner() and
handle all the planning.
For router planner, standard_planner() is mostly important to generate
the necessary restriction information. Later, the restriction information
generated by the standard_planner is used to decide whether all the shards
that a distributed query touches reside on a single worker node. However,
standard_planner() does a lot of extra things such as cost estimation and
execution path generations which are completely unnecessary in the context
of distributed planning.
There are certain types of queries where Citus could skip relying on
standard_planner() to generate the restriction information. For queries
in the following format, Citus does not need any information that the
standard_planner() generates:
SELECT ... FROM single_table WHERE distribution_key = X; or
DELETE FROM single_table WHERE distribution_key = X; or
UPDATE single_table SET value_1 = value_2 + 1 WHERE distribution_key = X;
Note that the queries might not be as simple as the above such that
GROUP BY, WINDOW FUNCIONS, ORDER BY or HAVING etc. are all acceptable. The
only rule is that the query is on a single distributed (or reference) table
and there is a "distribution_key = X;" in the WHERE clause. With that, we
could use to decide the shard that a distributed query touches reside on
a worker node.
We used to error out if there is a reference table
in the query participating a union. This has caused
pushdownable queries to be evaluated in coordinator.
Now we let reference tables inside union queries as long
as there is a distributed table in from clause.
Existing join checks (reference table on the outer part)
sufficient enought that we do not need check the join relation
of reference tables.
Previously we allowed task assignment policy to have affect on router queries
with only intermediate results. However, that is erroneous since the code-path
that assigns placements relies on shardIds and placements, which doesn't exists
for intermediate results.
With this commit, we do not apply task assignment policies when a router query
hits only intermediate results.
We update column attributes of various clauses for a query
inluding target columns, select clauses when we introduce
new range table entries in the query.
It seems having clause column attributes were not updated.
This fix resolves the issue
Before this commit, Citus supported INSERT...SELECT queries with
ON CONFLICT or RETURNING clauses only for pushdownable ones, since
queries supported via coordinator were utilizing COPY infrastructure
of PG to send selected tuples to the target worker nodes.
After this PR, INSERT...SELECT queries with ON CONFLICT or RETURNING
clauses will be performed in two phases via coordinator. In the first
phase selected tuples will be saved to the intermediate table which
is colocated with target table of the INSERT...SELECT query. Note that,
a utility function to save results to the colocated intermediate result
also implemented as a part of this commit. In the second phase, INSERT..
SELECT query is directly run on the worker node using the intermediate
table as the source table.
Description: Support round-robin `task_assignment_policy` for queries to reference tables.
This PR allows users to query multiple placements of shards in a round robin fashion. When `citus.task_assignment_policy` is set to `'round-robin'` the planner will use a round robin scheduling feature when multiple shard placements are available.
The primary use-case is spreading the load of reference table queries to all the nodes in the cluster instead of hammering only the first placement of the reference table. Since reference tables share the same path for selecting the shards with single shard queries that have multiple placements (`citus.shard_replication_factor > 1`) this setting also allows users to spread the query load on these shards.
For modifying queries we do not apply a round-robin strategy. This would be negated by an extra reordering step in the executor for such queries where a `first-replica` strategy is enforced.
The file handling the utility functions (DDL) for citus organically grew over time and became unreasonably large. This refactor takes that file and refactored the functionality into separate files per command. Initially modeled after the directory and file layout that can be found in postgres.
Although the size of the change is quite big there are barely any code changes. Only one two functions have been added for readability purposes:
- PostProcessIndexStmt which is extracted from PostProcessUtility
- PostProcessAlterTableStmt which is extracted from multi_ProcessUtility
A README.md has been added to `src/backend/distributed/commands` describing the contents of the module and every file in the module.
We need more documentation around the overloading of the COPY command, for now the boilerplate has been added for people with better knowledge to fill out.
PG 11 has change the way that PARAM_EXTERN is processed.
This commit ensures that Citus follows the same pattern.
For details see the related Postgres commit:
6719b238e8
Both of these are a bit of a shot in the dark. In one case, we noticed
a stack trace where a caller received a null pointer and attempted to
dereference the memory context field (at 0x010). In the other, I saw
that any error thrown from within AdjustParseTree could keep the stack
from being cleaned up (presumably if we push we should always pop).
Both stack traces were collected during times of high memory pressure
and locally reproducing the problem locally or otherwise has been very
tricky (i.e. it hasn't been reproduced reliably at all).
With this commit, we all partitioned distributed tables with
replication factor > 1. However, we also have many restrictions.
In summary, we disallow all kinds of modifications (including DDLs)
on the partition tables. Instead, the user is allowed to run the
modifications over the parent table.
The necessity for such a restriction have two aspects:
- We need to acquire shard resource locks appropriately
- We need to handle marking partitions INVALID in case
of any failures. Note that, in theory, the parent table
should also become INVALID, which is too aggressive.
This commit uses *_walker instead of *_mutator for performance reasons.
Given that we're only updating a functionId in the tree, the approach
seems fine.
This commit by default enables hiding shard names on MX workers
by simple replacing `pg_table_is_visible()` calls with
`citus_table_is_visible()` calls on the MX worker nodes. The latter
function filters out tables that are known to be shards.
The main motivation of this change is a better UX. The functionality
can be opted out via a GUC.
We also added two views, namely citus_shards_on_worker and
citus_shard_indexes_on_worker such that users can query
them to see the shards and their corresponding indexes.
We also added debug messages such that the filtered tables can
be interactively seen by setting the level to DEBUG1.
We can now support more complex count distinct operations by
pulling necessary columns to coordinator and evalutating the
aggreage at coordinator.
It supports broad range of expression with the restriction that
the expression must contain a column.
* Change worker_hash_partition_table() such that the
divergence between Citus planner's hashing and
worker_hash_partition_table() becomes the same.
* Rename single partitioning to single range partitioning.
* Add single hash repartitioning. Basically, logical planner
treats single hash and range partitioning almost equally.
Physical planner, on the other hand, treats single hash and
dual hash repartitioning almost equally (except for JoinPruning).
* Add a new GUC to enable this feature
This commit doesn't change any of the logic at all.
Instead, the goal is to:
* Get rid of any code duplication
* Incremental changes to the optimizer made it slightly hard
to follow the code, improve that and make it easier to
implement new features
* Simplify the code by moving each part of query processing (e.g.,
DISTINCT, LIMIT etc) into its own function
* Make the interaction between each part of the query more
obvious (e.g., How DISTINCT affects LIMIT etc)
- changes in ruleutils_11.c is reflected
- vacuum statement api change is handled. We now allow
multi-table vacuum commands.
- some other function header changes are reflected
- api conflicts between PG11 and earlier versions
are handled by adding shims in version_compat.h
- various regression tests are fixed due output and
functionality in PG1
- no change is made to support new features in PG11
they need to be handled by new commit
Before this commit, we had code duplication in the
WorkerExtendedOpNode(). The duplication was
noticeable and any change is prone to bugs.
The PR consists of 4 commits. Each commit incrementally
fixes the problem by moving certain parts of the duplicated
code into smaller, better-documented functions.
Before this commit, we had a divergence among
the creation of master/worker extended op nodes.
This commit moves the related parts into a single place
and allows the creation of master/extended op nodes to
share a common data structure.
PostgreSQL might remove some of the subqueries when they do not
contribute to the query result at all. Citus should not try to
access such subqueries during planning.
This PR adds support for multiple AND expressions in Having
for pushdown planner. We simply make a call to make_ands_explicit
from MultiLogicalPlanOptimize for the having qual in
workerExtendedOpNode.
After this commit large_table_shard_count wont be used to
check whether broadcast join, which is renamed as reference
join, can be applied. Reference join can only be applied over
reference tables.
After this change all the logic related to shard data fetch logic
will be removed. Planner won't plan any ShardFetchTask anymore.
Shard fetch related steps in real time executor and task-tracker
executor have been removed.
Pushing down limit and order by into workers may produce
wrong output when distinct on() clause has expressions,
aggregates, or window functions.
This checking allows pushing down of limits only if
distinct clause is a superset of group by clause. i.e. it contains all clauses in group by.
This is the first of series of window function work.
We can now support window functions that can be pushed down to workers.
Window function must have distribution column in the partition clause
to be pushed down.
We push down order by to worker query when limit is specified
(with some other additional checks). If the query has an expression
on an aggregate or avg aggregate by itself, and there is an order
by on this particular target we may send wrong order by to worker
query with potential to affect query result.
The fix creates a auxilary target entry in the worker query and
uses that target entry for sorting.
Before this PR, we were trusting on the columns of group by about
guaranteeing the uniqueness of the results. However, this assumption
is correct only if the columns in the group by is subset of columns
in the distinct clause. It can be wrong if we have part of group by
columns and some aggregation columns in the distinct clause. With
this PR, we add distinct plan on top of aggregate plan when necessary.
With #1804 (and related PRs), Citus gained the ability to
plan subqueries that are not safe to pushdown.
There are two high-level requirements for pushing down subqueries:
* Individual subqueries that require a merge step (i.e., GROUP BY
on non-distribution key, or LIMIT in the subquery etc). We've
handled such subqueries via #1876.
* Combination of subqueries that are not joined on distribution keys.
This commit aims to recursively plan some of such subqueries to make
the whole query safe to pushdown.
The main logic behind non colocated subquery joins is that we pick
an anchor range table entry and check for distribution key equality
of any other subqueries in the given query. If for a given subquery,
we cannot find distribution key equality with the anchor rte, we
recursively plan that subquery.
We also used a hacky solution for picking relations as the anchor range
table entries. The hack is that we wrap them into a subquery. This is only
necessary since some of the attribute equivalance checks are based on
queries rather than range table entries.
We used to only support pushdownable set operations inside a
subquery, however, we could easily expand the restriction
checks to cover top level set operations as well.
We use PostgreSQL hooks to accumulate the join restrictions
and PostgreSQL gives us all the join paths it tries while
deciding on the join order. Thus, for queries that have many
joins, this function is likely to remove lots of duplicate join
restrictions. This becomes relevant for Citus on query pushdown
check peformance.
We were allowing count distict queries even if they were
not directly on columns if the query is grouped on
distribution column.
When performing these checks we were skipping subqueries
because they also perform this check in a more concise manner.
We relied on oid SUBQUERY_RELATION_ID (10000) to decide if
a given RTE relation id denotes a subquery, however, we also
use SUBQUERY_PUSHDOWN_RELATION_ID (10001) for some subqueries.
We skip both type of subqueries with this change.
clause is not supported
This change allows unsupported clauses to go through query pushdown
planner instead of erroring out as we already do for non-outer joins.
We used to error out if the join clause includes filters like
t1.a < t2.a even if other filter like t1.key = t2.key exists.
Recently we lifted that restriction in subquery planning by
not lifting that restriction and focusing on equivalance classes
provided by postgres.
This checkin forwards previously erroring out real-time queries
due to join clauses to subquery planner and let it handle the
join even if the query does not have a subquery.
We are now pushing down queries that do not have any
subqueries in it. Error message looked misleading, changed to a more descriptive one.
We were creating intermediate query result's target
names from subquery target list. Now we also check
if cte re-defines its column name aliases, and create
intermediate result query accordingly.
With this commit, Citus recursively plans subqueries that
are not safe to pushdown, in other words, requires a merge
step.
The algorithm is simple: Recursively traverse the query from bottom
up (i.e., bottom meaning the leaf queries). On each level, check
whether the query is safe to pushdown (or a single repartition
subquery). If the answer is yes, do not touch that subquery. If the
answer is no, plan the subquery seperately (i.e., create a subPlan
for it) and replace the subquery with a call to
`read_intermediate_results(planId, subPlanId)`. During the the
execution, run the subPlans first, and make them avaliable to the
next query executions.
Some of the queries hat this change allows us:
* Subqueries with LIMIT
* Subqueries with GROUP BY/DISTINCT on non-partition keys
* Subqueries involving re-partition joins, router queries
* Mixed usage of subqueries and CTEs (i.e., use CTEs in
subqueries as well). Nested subqueries as long as we
support the subquery inside the nested subquery.
* Subqueries with local tables (i.e., those subqueries
has the limitation that they have to be leaf subqueries)
* VIEWs on the distributed tables just works (i.e., the
limitations mentioned below still applies to views)
Some of the queries that is still NOT supported:
* Corrolated subqueries that are not safe to pushdown
* Window function on non-partition keys
* Recursively planned subqueries or CTEs on the outer
side of an outer join
* Only recursively planned subqueries and CTEs in the FROM
(i.e., not any distributed tables in the FROM) and subqueries
in WHERE clause
* Subquery joins that are not on the partition columns (i.e., each
subquery is individually joined on partition keys but not the upper
level subquery.)
* Any limitation that logical planner applies such as aggregate
distincts (except for count) when GROUP BY is on non-partition key,
or array_agg with ORDER BY
Postgres provides OS agnosting formatting macros for
formatting 64 bit numbers. Replaced %ld %lu with
INT64_FORMAT and UINT64_FORMAT respectively.
Also found some incorrect usages of formatting
flags and fixed them.
We added the ability to filter out the planner restriction information
for specific parts of the query. This might lead to situations where
the common restriction includes some other relations that we're searching
for. The reason is that while filtering for join restrictions, we add the
restriction as soon as we find the relation.
With this commit we make sure that the common attribute
equivalance class always includes the input relations.
In subquery pushdown, we first ensure that each relation is joined with at least
on another relation on the partition keys. That's fine given that the decision
is binary: pushdown the query at all or not.
With recursive planning, we'd want to check whether any specific part
of the query can be pushded down or not. Thus, we need the ability to
understand which part(s) of the subquery is safe to pushdown. This commit
adds the infrastructure for doing that.
Note that we used to iterate over the RTEs once for performance reasons.
However, keeping an extra copy of original query seems more costly and
hard to maintain/explain.
Subquery pushdown planning is based on relation restriction
equivalnce. This brings us the opportuneatly to allow any
other joins as long as there is an already equi join between
the distributed tables.
We already allow that for joins with reference tables and
this commit allows that for joins among distributed tables.
With this commit, we allow pushing down subqueries with only
reference tables where GROUP BY or DISTINCT clause or Window
functions include only columns from reference tables.
It's possible to build INSERT SELECT queries which include implicit
casts, currently we attempt to support these by adding explicit casts to
the SELECT query, but this sometimes crashes because we don't update all
nodes with the new types. (SortClauses, for instance)
This commit removes those explicit casts and passes an unmodified SELECT
query to the COPY executor (how we implement INSERT SELECT under the
scenes). In lieu of those cases, COPY has been given some extra logic to
inspect queries, notice that the types don't line up with the table it's
supposed to be inserting into, and "manually" casting every tuple before
sending them to workers.
Expands count distinct coverage by allowing more cases. We used to support
count distinct only if we can push down distinct aggregate to worker query
i.e. the count distinct clause was on the partition column of the table,
or there was a grouping on the partition column.
Now we can support
- non-partition columns, with or without grouping on partition column
- partition, and non partition column in the same query
- having clause
- single table subqueries
- insert into select queries
- join queries where count distinct is on partition, or non-partition column
- filters on count distinct clauses (extends existing support)
We first try to push down aggregate to worker query (original case), if we
can't then we modify worker query to return distinct columns to coordinator
node. We do that by adding distinct column targets to group by clauses. Then
we perform count distinct operation on the coordinator node.
This work should reduce the cases where HLL is used as it can address anything
that HLL can. However, if we start having performance issues due to very large
number rows, then we can recommend hll use.
This commit provides the support for window functions in subquery and insert
into select queries. Note that our support for window functions is still limited
because it must have a partition by clause on the distribution key. This commit
makes changes in the files insert_select_planner and multi_logical_planner. The
required tests are also added with files multi_subquery_window_functions.out
and multi_insert_select_window.out.
Citus can handle INSERT INTO ... SELECT queries if the query inserts
into local table by reading data from distributed table. The opposite
way is not correct. With this commit we warn the user if the latter
option is used.
With this commit, we relax the restrictions put on the reference
tables with subquery pushdown.
We did three notable improvements:
1) Relax equi-join restrictions
Previously, we always expected that the non-reference tables are
equi joined with reference tables on the partition key of the
non-reference table.
With this commit, we allow any column of non-reference tables
joined using non-equi joins as well.
2) Relax OUTER JOIN restrictions
Previously Citus errored out if any reference table exists at
any point of the outer part of an outer join. For instance,
See the below sketch where (h) denotes a hash distributed relation,
(r) denotes a reference table, (L) denotes LEFT JOIN and
(I) denotes INNER JOIN.
(L)
/ \
(I) h
/ \
r h
Before this commit Citus would error out since a reference table
appears on the left most part of an left join. However, that was
too restrictive so that we only error out if the reference table
is directly below and in the outer part of an outer join.
3) Bug fixes
We've done some minor bugfixes in the existing implementation.
This is necessary for multi-row INSERTs for the same reasons we use it
in e.g. UPSERTs: if the range table list has more than one entry, then
PostgreSQL's deparse logic requires that vars be prefixed by the name
of their corresponding range table entry. This of course doesn't affect
single-row INSERTs, but since multi-row INSERTs have a VALUE RTE, they
were affected.
The piece of ruleutils which builds range table names wasn't modified
to handle shard extension; instead UPSERT/INSERT INTO ... SELECT added
an alias to the RTE. When present, this alias is favored. Doing the
same in the multi-row INSERT case fixes RETURNING for such commands.
This GUC has two settings, 'always' and 'never'. When it's set to
'never' all behavior stays exactly as it was prior to this commit. When
it's set to 'always' only SELECT queries are allowed to run, and only
secondary nodes are used when processing those queries.
Add some helper functions:
- WorkerNodeIsSecondary(), checks the noderole of the worker node
- WorkerNodeIsReadable(), returns whether we're currently allowed to
read from this node
- ActiveReadableNodeList(), some functions (namely, the ones on the
SELECT path) don't require working with Primary Nodes. They should call
this function instead of ActivePrimaryNodeList(), because the latter
will error out in contexts where we're not allowed to write to nodes.
- ActiveReadableNodeCount(), like the above, replaces
ActivePrimaryNodeCount().
- EnsureModificationsCanRun(), error out if we're not currently allowed
to run queries which modify data. (Either we're in read-only mode or
use_secondary_nodes is set)
Some parts of the code were switched over to use readable nodes instead
of primary nodes:
- Deadlock detection
- DistributedTableSize,
- the router, real-time, and task tracker executors
- ShardPlacement resolution
This is a pretty substantial refactoring of the existing modify path
within the router executor and planner. In particular, we now hunt for
all VALUES range table entries in INSERT statements and group the rows
contained therein by shard identifier. These rows are stashed away for
later in "ModifyRoute" elements. During deparse, the appropriate RTE
is extracted from the Query and its values list is replaced by these
rows before any SQL is generated.
In this way, we can create multiple Tasks, but only one per shard, to
piecemeal execute a multi-row INSERT. The execution of jobs containing
such tasks now exclusively go through the "multi-router executor" which
was previously used for e.g. INSERT INTO ... SELECT.
By piggybacking onto that executor, we participate in ongoing trans-
actions, get rollback-ability, etc. In short order, the only remaining
use of the "single modify" router executor will be for bare single-
row INSERT statements (i.e. those not in a transaction).
This change appropriately handles deferred pruning as well as master-
evaluated functions.
In standart_planner, PostgreSQL expands partitioned tables to their
partitions and call our restriction hook for each partition. It also,
for some queries, skips the partitioned table itself completely. This
behaviour makes it difficult to prune shards and decide whether query
is router plannable or not. To prevent this behaviour, we change inh
flag of partitioned tables to false in the query tree. In this case,
PostgreSQL treats those partitioned tables as regular relations and
does not expand them.
This behaviour is inline with our expectations, because we do not want
to treat partitioned tables differently on coordinator. Although we are
not entirely comfortable with modifying query tree, other solutions to
this problem is overly complicated.
- master_add_node enforces that there is only one primary per group
- there's also a trigger on pg_dist_node to prevent multiple primaries
per group
- functions in metadata cache only return primary nodes
- Rename ActiveWorkerNodeList -> ActivePrimaryNodeList
- Rename WorkerGetLive{Node->Group}Count()
- Refactor WorkerGetRandomCandidateNode
- master_remove_node only complains about active shard placements if the
node being removed is a primary.
- master_remove_node only deletes all reference table placements in the
group if the node being removed is the primary.
- Rename {Node->NodeGroup}HasShardPlacements, this reflects the behavior it
already had.
- Rename DeleteAllReferenceTablePlacementsFrom{Node->NodeGroup}. This also
reflects the behavior it already had, but the new signature forces the
caller to pass in a groupId
- Rename {WorkerGetLiveGroup->ActivePrimaryNode}Count
Comes with a few changes:
- Change the signature of some functions to accept groupid
- InsertShardPlacementRow
- DeleteShardPlacementRow
- UpdateShardPlacementState
- NodeHasActiveShardPlacements returns true if the group the node is a
part of has any active shard placements
- TupleToShardPlacement now returns ShardPlacements which have NULL
nodeName and nodePort.
- Populate (nodeName, nodePort) when creating ShardPlacements
- Disallow removing a node if it contains any shard placements
- DeleteAllReferenceTablePlacementsFromNode matches based on group. This
doesn't change behavior for now (while there is only one node per
group), but means in the future callers should be careful about
calling it on a secondary node, it'll delete placements on the primary.
- Create concept of a GroupShardPlacement, which represents an actual
tuple in pg_dist_placement and is distinct from a ShardPlacement,
which has been resolved to a specific node. In the future
ShardPlacement should be renamed to NodeShardPlacement.
- Create some triggers which allow existing code to continue to insert
into and update pg_dist_shard_placement as if it still existed.
Uncrustify 0.65 appears to have changed some defaults, resulting in
breakages for those of us who have already upgraded; Travis still uses
Uncrustify 0.64, but these changes work with both versions (assuming
appropriately updated config), so this should permit use of either
version for the time being.
MasterIrreducibleExpressionWalker has a copied code from
function check_functions_in_node() which was available with
PG 9.6+. Now PG 9.5 support is dropped we can remove
duplicate code and directly call check_functions_in_node().
Adds support for PostgreSQL 10 by copying in the requisite ruleutils
and updating all API usages to conform with changes in PostgreSQL 10.
Most changes are fairly minor but they are numerous. One particular
obstacle was the change in \d behavior in PostgreSQL 10's psql; I had
to add SQL implementations (views, mostly) to mimic the pre-10 output.
Add a second implementation of INSERT INTO distributed_table SELECT ... that is used if
the query cannot be pushed down. The basic idea is to execute the SELECT query separately
and pass the results into the distributed table using a CopyDestReceiver, which is also
used for COPY and create_distributed_table. When planning the SELECT, we go through
planner hooks again, which means the SELECT can also be a distributed query.
EXPLAIN is supported, but EXPLAIN ANALYZE is not because preventing double execution was
a lot more complicated in this case.
Distributed query planning for subquery pushdown is done on the original
query. This prevents the usage of external parameters on the execution.
To overcome this, we manually replace the parameters on the original
query.
* Support for subqueries in WHERE clause
This commit enables subqueries in WHERE clause to be pushed down
by the subquery pushdown logic.
The support covers:
- Correlated subqueries with IN, NOT IN, EXISTS, NOT EXISTS,
operator expressions such as (>, <, =, ALL, ANY etc.)
- Non-correlated subqueries with (partition_key) IN (SELECT partition_key ..)
(partition_key) =ANY (SELECT partition_key ...)
Note that this commit heavily utilizes the attribute equivalence logic introduced
in the 1cb6a34ba8. In general, this commit mostly
adjusts the logical planner not to error out on the subqueries in WHERE clause.
* Improve error checks for subquery pushdown and INSERT ... SELECT
Since we allow subqueries in WHERE clause with the previous commit,
we should apply the same limitations to those subqueries.
With this commit, we do not iterate on each subquery one by one.
Instead, we extract all the subqueries and apply the checks directly
on those subqueries. The aim of this change is to (i) Simplify the
code (ii) Make it close to the checks on INSERT .. SELECT code base.
* Extend checks for unresolved paramaters to include SubLinks
With the presence of subqueries in where clause (i.e., SubPlans on the
query) the existing way for checking unresolved parameters fail. The
reason is that the parameters for SubPlans are kept on the parent plan not
on the query itself (see primnodes.h for the details).
With this commit, instead of checking SubPlans on the modified plans
we start to use originalQuery, where SubLinks represent the subqueries
in where clause. The unresolved parameters can be found on the SubLinks.
* Apply code-review feedback
* Remove unnecessary copying of shard interval list
This commit removes unnecessary copying of shard interval list. Note
that there are no copyObject function implemented for shard intervals.
* Enabling physical planner for subquery pushdown changes
This commit applies the logic that exists in INSERT .. SELECT
planning to the subquery pushdown changes.
The main algorithm is followed as :
- pick an anchor relation (i.e., target relation)
- per each target shard interval
- add the target shard interval's shard range
as a restriction to the relations (if all relations
joined on the partition keys)
- Check whether the query is router plannable per
target shard interval.
- If router plannable, create a task
* Add union support within the JOINS
This commit adds support for UNION/UNION ALL subqueries that are
in the following form:
.... (Q1 UNION Q2 UNION ...) as union_query JOIN (QN) ...
In other words, we currently do NOT support the queries that are
in the following form where union query is not JOINed with
other relations/subqueries :
.... (Q1 UNION Q2 UNION ...) as union_query ....
* Subquery pushdown planner uses original query
With this commit, we change the input to the logical planner for
subquery pushdown. Before this commit, the planner was relying
on the query tree that is transformed by the postgresql planner.
After this commit, the planner uses the original query. The main
motivation behind this change is the simplify deparsing of
subqueries.
* Enable top level subquery join queries
This work enables
- Top level subquery joins
- Joins between subqueries and relations
- Joins involving more than 2 range table entries
A new regression test file is added to reflect enabled test cases
* Add top level union support
This commit adds support for UNION/UNION ALL subqueries that are
in the following form:
.... (Q1 UNION Q2 UNION ...) as union_query ....
In other words, Citus supports allow top level
unions being wrapped into aggregations queries
and/or simple projection queries that only selects
some fields from the lower level queries.
* Disallow subqueries without a relation in the range table list for subquery pushdown
This commit disallows subqueries without relation in the range table
list. This commit is only applied for subquery pushdown. In other words,
we do not add this limitation for single table re-partition subqueries.
The reasoning behind this limitation is that if we allow pushing down
such queries, the result would include (shardCount * expectedResults)
where in a non distributed world the result would be (expectedResult)
only.
* Disallow subqueries without a relation in the range table list for INSERT .. SELECT
This commit disallows subqueries without relation in the range table
list. This commit is only applied for INSERT.. SELECT queries.
The reasoning behind this limitation is that if we allow pushing down
such queries, the result would include (shardCount * expectedResults)
where in a non distributed world the result would be (expectedResult)
only.
* Change behaviour of subquery pushdown flag (#1315)
This commit changes the behaviour of the citus.subquery_pushdown flag.
Before this commit, the flag is used to enable subquery pushdown logic. But,
with this commit, that behaviour is enabled by default. In other words, the
flag is now useless. We prefer to keep the flag since we don't want to break
the backward compatibility. Also, we may consider using that flag for other
purposes in the next commits.
* Require subquery_pushdown when limit is used in subquery
Using limit in subqueries may cause returning incorrect
results. Therefore we allow limits in subqueries only
if user explicitly set subquery_pushdown flag.
* Evaluate expressions on the LIMIT clause (#1333)
Subquery pushdown uses orignal query, the LIMIT and OFFSET clauses
are not evaluated. However, logical optimizer expects these expressions
are already evaluated by the standard planner. This commit manually
evaluates the functions on the logical planner for subquery pushdown.
* Better format subquery regression tests (#1340)
* Style fix for subquery pushdown regression tests
With this commit we intented a more consistent style for the
regression tests we've added in the
- multi_subquery_union.sql
- multi_subquery_complex_queries.sql
- multi_subquery_behavioral_analytics.sql
* Enable the tests that are temporarily commented
This commit enables some of the regression tests that were commented
out until all the development is done.
* Fix merge conflicts (#1347)
- Update regression tests to meet the changes in the regression
test output.
- Replace Ifs with Asserts given that the check is already done
- Update shard pruning outputs
* Add view regression tests for increased subquery coverage (#1348)
- joins between views and tables
- joins between views
- union/union all queries involving views
- views with limit
- explain queries with view
* Improve btree operators for the subquery tests
This commit adds the missing comprasion for subquery composite key
btree comparator.
We previously dismissed this as unimportant, but it turns out to be
very useful for the upcoming subquery pushdown, where a user might
specify an equality constraint in a subquery, and the subquery
pushdown machinery adds >= and <= restrictions on the shard boundary.
Previously the latter restriction was ignored.
It semms that GEQO optimizations, when it is set to on, create their own memory context
and free it after when it is no longer necessary. In join multi_join_restriction_hook
we allocate our variables in the CurrentMemoryContext, which is GEQO's memory context
if it is active. To prevent deallocation of our variables when GEQO's memory context is
freed, we started to allocate memory fo these variables in separate MemoryContext.
So far citus used postgres' predicate proofing logic for shard
pruning, except for INSERT and COPY which were already optimized for
speed. That turns out to be too slow:
* Shard pruning for SELECTs is currently O(#shards), because
PruneShardList calls predicate_refuted_by() for every
shard. Obviously using an O(N) type algorithm for general pruning
isn't good.
* predicate_refuted_by() is quite expensive on its own right. That's
primarily because it's optimized for doing a single refutation
proof, rather than performing the same proof over and over.
* predicate_refuted_by() does not keep persistent state (see 2.) for
function calls, which means that a lot of syscache lookups will be
performed. That's particularly bad if the partitioning key is a
composite key, because without a persistent FunctionCallInfo
record_cmp() has to repeatedly look-up the type definition of the
composite key. That's quite expensive.
Thus replace this with custom-code that works in two phases:
1) Search restrictions for constraints that can be pruned upon
2) Use those restrictions to search for matching shards in the most
efficient manner available:
a) Binary search / Hash Lookup in case of hash partitioned tables
b) Binary search for equal clauses in case of range or append
tables without overlapping shards.
c) Binary search for inequality clauses, searching for both lower
and upper boundaries, again in case of range or append
tables without overlapping shards.
d) exhaustive search testing each ShardInterval
My measurements suggest that we are considerably, often orders of
magnitude, faster than the previous solution, even if we have to fall
back to exhaustive pruning.
With this commit, we started to send explain queries within a savepoint. After
running explain query, we rollback to savepoint. This saves us from side effects
of EXPLAIN ANALYZE on DML queries.
All callers fetch a cache entry and extract/compute arguments for the
eventual FindShardInterval call, so it makes more sense to refactor
into that function itself; this solves the use-after-free bug, too.
Soon shard pruning will be optimized not to generally work linearly
anymore. Thus we can't print the pruned shard intervals as currently
done anymore.
The current printing of shard ids also prevents us from running tests
in parallel, as otherwise shard ids aren't linearly numbered.
With this change we add an option to add a node without replicating all reference
tables to that node. If a node is added with this option, we mark the node as
inactive and no queries will sent to that node.
We also added two new UDFs;
- master_activate_node(host, port):
- marks node as active and replicates all reference tables to that node
- master_add_inactive_node(host, port):
- only adds node to pg_dist_node
Before this commit, we were erroring out for queries containing parameterized SQL functions
like 'SELECT parameterized_sql_query(value)' as we should, however we were returning wrong
results for queries like 'SELECT * FROM parameterized_sql_query(value)'. With this commit
we started to error out on such queries too.
In this PR, we aim to deduce whether each of the RTE_RELATION
is joined with at least on another RTE_RELATION on their partition keys. If each
RTE_RELATION follows the above rule, we can conclude that all RTE_RELATIONs are
joined on their partition keys.
In order to do that, we invented a new equivalence class namely:
AttributeEquivalenceClass. In very simple words, a AttributeEquivalenceClass is
identified by an unique id and consists of a list of AttributeEquivalenceMembers.
Each AttributeEquivalenceMember is designed to identify attributes uniquely within the
whole query. The necessity of this arise since varno attributes are defined within
a single level of a query. Instead, here we want to identify each RTE_RELATION uniquely
and try to find equality among each RTE_RELATION's partition key.
Whenever we find an equality clause A = B, where both A and B originates from
relation attributes (i.e., not random expressions), we create an
AttributeEquivalenceClass to record this knowledge. If we later find another
equivalence B = C, we create another AttributeEquivalenceClass. Finally, we can
apply transitity rules and generate a new AttributeEquivalenceClass which includes
A, B and C.
Note that equality among the members are identified by the varattno and rteIdentity.
Each equality among RTE_RELATION is saved using an AttributeEquivalenceClass where
each member attribute is identified by a AttributeEquivalenceMember. In the final
step, we try generate a common attribute equivalence class that holds as much as
AttributeEquivalenceMembers whose attributes are a partition keys.
Before this commit, in certain cases router planner allowed pushing
down JOINs that are not on the partition keys.
With @anarazel's suggestion, we change the logic to use uninstantiated
parameter. Previously, the planner was traversing on the restriction
information and once it finds the parameter, it was replacing it with
the shard range. With this commit, instead of traversing the restrict
infos, the planner explicitly checks for the equivalence of the relation
partition key with the uninstantiated parameter. If finds an equivalence,
it adds the restrictions. In this way, we have more control over the
queries that are pushed down.
With this commit, we add the range table list of the original query to our
custom plan. Therefore, PostgreSQL can check relations in the original query
for access permissions and error out if the proper access is not granted.
Custom Scan is a node in the planned statement which helps external providers
to abstract data scan not just for foreign data wrappers but also for regular
relations so you can benefit your version of caching or hardware optimizations.
This sounds like only an abstraction on the data scan layer, but we can use it
as an abstraction for our distributed queries. The only thing we need to do is
to find distributable parts of the query, plan for them and replace them with
a Citus Custom Scan. Then, whenever PostgreSQL hits this custom scan node in
its Vulcano style execution, it will call our callback functions which run
distributed plan and provides tuples to the upper node as it scans a regular
relation. This means fewer code changes, fewer bugs and more supported features
for us!
First, in the distributed query planner phase, we create a Custom Scan which
wraps the distributed plan. For real-time and task-tracker executors, we add
this custom plan under the master query plan. For router executor, we directly
pass the custom plan because there is not any master query. Then, we simply let
the PostgreSQL executor run this plan. When it hits the custom scan node, we
call the related executor parts for distributed plan, fill the tuple store in
the custom scan and return results to PostgreSQL executor in Vulcano style,
a tuple per XXX_ExecScan() call.
* Modify planner to utilize Custom Scan node.
* Create different scan methods for different executors.
* Use native PostgreSQL Explain for master part of queries.
During later work the transaction debug output will change (as it will
in postgres 10), which makes it hard to see actual changes in the
INSERT ... SELECT ... test. Reduce to DEBUG2 after changing a debug
message to that log level.
All router, real-time, task-tracker plannable queries should now have
full prepared statement support (and even use router when possible),
unless they don't go through the custom plan interface (which
basically just affects LANGUAGE SQL (not plpgsql) functions).
This is achieved by forcing postgres' planner to always choose a
custom plan, by assigning very low costs to plans with bound
parameters (i.e. ones were the postgres planner replanned the query
upon EXECUTE with all parameter values provided), instead of the
generic one.
This requires some trickery, because for custom plans to work the
costs for a non-custom plan have to be known, which means we can't
error out when planning the generic plan. Instead we have to return a
"faux" plan, that'd trigger an error message if executed. But due to
the custom plan logic that plan will likely (unless called by an SQL
function, or because we can't support that query for some reason) not
be executed; instead the custom plan will be chosen.
So far router planner had encapsulated different functionality in
MultiRouterPlanCreate. Modifications always go through router, selects
sometimes. Modifications always error out if the query is unsupported,
selects return NULL. Especially the error handling is a problem for
the upcoming extension of prepared statement support.
Split MultiRouterPlanCreate into CreateRouterPlan and
CreateModifyPlan, and change them to not throw errors.
Instead errors are now reported by setting the new
MultiPlan->plannigError.
Callers of router planner functionality now have to throw errors
themselves if desired, but also can skip doing so.
This is a pre-requisite for expanding prepared statement support.
While touching all those lines, improve a number of error messages by
getting them closer to the postgres error message guidelines.
The name CreatePhysicalPlan() hasn't been accurate for a while, and
the split of work between multi_planner() and CreatePhysicalPlan()
doesn't seem perfect. So rename to CreateDistributedPlan() and move a
bit more logic in there.
This adds a replication_model GUC which is used as the replication
model for any new distributed table that is not a reference table.
With this change, tables with replication factor 1 are no longer
implicitly MX tables.
The GUC is similarly respected during empty shard creation for e.g.
existing append-partitioned tables. If the model is set to streaming
while replication factor is greater than one, table and shard creation
routines will error until this invalid combination is corrected.
Changing this parameter requires superuser permissions.
We changed error message which appears when user tries to execute outer join command and
that command requires repartitioning. Old error message mentioned about 1-to-1 shard
partitioning which may not be clear to user.
Instead use pg_plan_query() like the normal explain does, and use that
to explain the query. That's important because it allows to remove
the duplicated planner logic from multi_explain - and that logic is
about to get more complicated.
They make fixing explain for prepared statement harder, and they don't
really fit into EXPLAIN in the first place. Additionally they're
currently not exercised in any tests.
This commit is intended to improve the error messages while planning
INSERT INTO .. SELECT queries. The main motivation for this change is
that we used to map multiple cases into a single message. With this change,
we added explicit error messages for many cases.
Enables use views within distributed queries.
User can create and use a view on distributed tables/queries
as he/she would use with regular queries.
After this change router queries will have full support for views,
insert into select queries will support reading from views, not
writing into. Outer joins would have a limited support, and would
error out at certain cases such as when a view is in the inner side
of the outer join.
Although PostgreSQL supports writing into views under certain circumstances.
We disallowed that for distributed views.
With this change we start to error out on router planner queries where a common table
expression with data-modifying statement is present. We already do not support if
there is a data-modifying statement using result of the CTE, now we also error out
if CTE itself is data-modifying statement.
SaitTalhaNisanci
Philip Dubé
Nils Dijk
Marco Slot
Onder Kalaci
Onur Tirtir
Hanefi Onaldi
Jelte Fennema
Hadi Moshayedi
Markus Sintonen
Önder Kalacı
Khashayar Fereidani
Philip Dubé
Hanefi Onaldi
Marco Slot
Hadi Moshayedi
exialin
Jason Petersen
Murat Tuncer
velioglu
Jason Petersen
Nils Dijk
Jason Petersen
mehmet furkan şahin
Murat Tuncer
Metin Doslu
Brian Cloutier
Hadi Moshayedi
Burak Yucesoy
Andres Freund
jmunsch
Eren Basak