Spracherkennung für: .out vermutete Sprache: Unknown {[0] [0] [0]} [Methode: Schwerpunktbildung, einfache Gewichte, sechs Dimensionen]
Parsed test spec with 2 sessions
starting permutation: b1 m1 s1 c1 b2 m2 s2 c2
step b1: BEGIN ISOLATION LEVEL READ COMMITTED;
step m1: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
step s1: SELECT * FROM tgt;
id|val
--+---
1| 10
2| 20
3| 30
(3 rows)
step c1: COMMIT;
step b2: BEGIN ISOLATION LEVEL READ COMMITTED;
step m2: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
step s2: SELECT * FROM tgt;
id|val
--+---
1| 10
2| 20
3| 30
(3 rows)
step c2: COMMIT;
starting permutation: b1 b2 m1 hj ex m2 c1 c2 s1
step b1: BEGIN ISOLATION LEVEL READ COMMITTED;
step b2: BEGIN ISOLATION LEVEL READ COMMITTED;
step m1: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
step hj: SET LOCAL enable_mergejoin = off; SET LOCAL enable_nestloop = off;
step ex: EXPLAIN (verbose, costs off)
MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
QUERY PLAN
---------------------------------------------------
Merge on public.tgt
-> Hash Left Join
Output: tgt.ctid, src.val, src.id, src.ctid
Inner Unique: true
Hash Cond: (src.id = tgt.id)
-> Seq Scan on public.src
Output: src.val, src.id, src.ctid
-> Hash
Output: tgt.ctid, tgt.id
-> Seq Scan on public.tgt
Output: tgt.ctid, tgt.id
(11 rows)
step m2: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val); <waiting ...>
step c1: COMMIT;
step m2: <... completed>
step c2: COMMIT;
step s1: SELECT * FROM tgt;
id|val
--+---
1| 10
2| 20
3| 30
(3 rows)
starting permutation: b1 b2 m1 mj ex m2 c1 c2 s1
step b1: BEGIN ISOLATION LEVEL READ COMMITTED;
step b2: BEGIN ISOLATION LEVEL READ COMMITTED;
step m1: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
step mj: SET LOCAL enable_hashjoin = off; SET LOCAL enable_nestloop = off;
step ex: EXPLAIN (verbose, costs off)
MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
QUERY PLAN
---------------------------------------------------
Merge on public.tgt
-> Merge Left Join
Output: tgt.ctid, src.val, src.id, src.ctid
Inner Unique: true
Merge Cond: (src.id = tgt.id)
-> Index Scan using src_pkey on public.src
Output: src.val, src.id, src.ctid
-> Index Scan using tgt_pkey on public.tgt
Output: tgt.ctid, tgt.id
(9 rows)
step m2: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val); <waiting ...>
step c1: COMMIT;
step m2: <... completed>
step c2: COMMIT;
step s1: SELECT * FROM tgt;
id|val
--+---
1| 10
2| 20
3| 30
(3 rows)
starting permutation: b1 b2 m1 nl ex m2 c1 c2 s1
step b1: BEGIN ISOLATION LEVEL READ COMMITTED;
step b2: BEGIN ISOLATION LEVEL READ COMMITTED;
step m1: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
step nl: SET LOCAL enable_hashjoin = off; SET LOCAL enable_mergejoin = off;
step ex: EXPLAIN (verbose, costs off)
MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val);
QUERY PLAN
---------------------------------------------------
Merge on public.tgt
-> Nested Loop Left Join
Output: tgt.ctid, src.val, src.id, src.ctid
Inner Unique: true
-> Seq Scan on public.src
Output: src.val, src.id, src.ctid
-> Index Scan using tgt_pkey on public.tgt
Output: tgt.ctid, tgt.id
Index Cond: (tgt.id = src.id)
(9 rows)
step m2: MERGE INTO tgt USING src ON tgt.id = src.id
WHEN MATCHED THEN UPDATE SET val = src.val
WHEN NOT MATCHED THEN INSERT VALUES (src.id, src.val); <waiting ...>
step c1: COMMIT;
step m2: <... completed>
step c2: COMMIT;
step s1: SELECT * FROM tgt;
id|val
--+---
1| 10
2| 20
3| 30
(3 rows)
[Dauer der Verarbeitung: 0.12 Sekunden, vorverarbeitet 2026-09-28]
