Avoid creation of the free space map for small heap relations.

Previously, all heaps had FSMs. For very small tables, this means that the
FSM took up more space than the heap did. This is wasteful, so now we
refrain from creating the FSM for heaps with 4 pages or fewer. If the last
known target block has insufficient space, we still try to insert into some
other page before giving up and extending the relation, since doing
otherwise leads to table bloat. Testing showed that trying every page
penalized performance slightly, so we compromise and try every other page.
This way, we visit at most two pages. Any pages with wasted free space
become visible at next relation extension, so we still control table bloat.
As a bonus, directly attempting one or two pages can even be faster than
consulting the FSM would have been.

Once the FSM is created for a heap we don't remove it even if somebody
deletes all the rows from the corresponding relation.  We don't think it is
a useful optimization as it is quite likely that relation will again grow
to the same size.

Author: John Naylor with design inputs and some code contribution by Amit Kapila
Reviewed-by: Amit Kapila
Tested-by: Mithun C Y
Discussion: https://www.postgresql.org/message-id/CAJVSVGWvB13PzpbLEecFuGFc5V2fsO736BsdTakPiPAcdMM5tQ@mail.gmail.com

1 files changed, 39 insertions, 38 deletions

diff --git a/contrib/pageinspect/expected/page.out b/contrib/pageinspect/expected/page.out
index 3fcd9fbe6d9..11d56f49a4b 100644
--- a/contrib/pageinspect/expected/page.out
+++ b/contrib/pageinspect/expected/page.out
@@ -1,48 +1,69 @@
 CREATE EXTENSION pageinspect;
-CREATE TABLE test1 (a int, b int);
-INSERT INTO test1 VALUES (16777217, 131584);
-VACUUM test1;  -- set up FSM
+CREATE TABLE test_rel_forks (a int);
+-- Make sure there are enough blocks in the heap for the FSM to be created.
+INSERT INTO test_rel_forks SELECT i from generate_series(1,1000) i;
+-- set up FSM and VM
+VACUUM test_rel_forks;
 -- The page contents can vary, so just test that it can be read
 -- successfully, but don't keep the output.
-SELECT octet_length(get_raw_page('test1', 'main', 0)) AS main_0;
+SELECT octet_length(get_raw_page('test_rel_forks', 'main', 0)) AS main_0;
  main_0 
 --------
    8192
 (1 row)
 
-SELECT octet_length(get_raw_page('test1', 'main', 1)) AS main_1;
-ERROR:  block number 1 is out of range for relation "test1"
-SELECT octet_length(get_raw_page('test1', 'fsm', 0)) AS fsm_0;
+SELECT octet_length(get_raw_page('test_rel_forks', 'main', 100)) AS main_100;
+ERROR:  block number 100 is out of range for relation "test_rel_forks"
+SELECT octet_length(get_raw_page('test_rel_forks', 'fsm', 0)) AS fsm_0;
  fsm_0 
 -------
   8192
 (1 row)
 
-SELECT octet_length(get_raw_page('test1', 'fsm', 1)) AS fsm_1;
- fsm_1 
--------
-  8192
-(1 row)
-
-SELECT octet_length(get_raw_page('test1', 'vm', 0)) AS vm_0;
+SELECT octet_length(get_raw_page('test_rel_forks', 'fsm', 10)) AS fsm_10;
+ERROR:  block number 10 is out of range for relation "test_rel_forks"
+SELECT octet_length(get_raw_page('test_rel_forks', 'vm', 0)) AS vm_0;
  vm_0 
 ------
  8192
 (1 row)
 
-SELECT octet_length(get_raw_page('test1', 'vm', 1)) AS vm_1;
-ERROR:  block number 1 is out of range for relation "test1"
+SELECT octet_length(get_raw_page('test_rel_forks', 'vm', 1)) AS vm_1;
+ERROR:  block number 1 is out of range for relation "test_rel_forks"
 SELECT octet_length(get_raw_page('xxx', 'main', 0));
 ERROR:  relation "xxx" does not exist
-SELECT octet_length(get_raw_page('test1', 'xxx', 0));
+SELECT octet_length(get_raw_page('test_rel_forks', 'xxx', 0));
 ERROR:  invalid fork name
 HINT:  Valid fork names are "main", "fsm", "vm", and "init".
-SELECT get_raw_page('test1', 0) = get_raw_page('test1', 'main', 0);
+SELECT * FROM fsm_page_contents(get_raw_page('test_rel_forks', 'fsm', 0));
+ fsm_page_contents 
+-------------------
+ 0: 147           +
+ 1: 147           +
+ 3: 147           +
+ 7: 147           +
+ 15: 147          +
+ 31: 147          +
+ 63: 147          +
+ 127: 147         +
+ 255: 147         +
+ 511: 147         +
+ 1023: 147        +
+ 2047: 147        +
+ 4095: 147        +
+ fp_next_slot: 0  +
+ 
+(1 row)
+
+SELECT get_raw_page('test_rel_forks', 0) = get_raw_page('test_rel_forks', 'main', 0);
  ?column? 
 ----------
  t
 (1 row)
 
+DROP TABLE test_rel_forks;
+CREATE TABLE test1 (a int, b int);
+INSERT INTO test1 VALUES (16777217, 131584);
 SELECT pagesize, version FROM page_header(get_raw_page('test1', 0));
  pagesize | version 
 ----------+---------
@@ -62,26 +83,6 @@ SELECT tuple_data_split('test1'::regclass, t_data, t_infomask, t_infomask2, t_bi
  {"\\x01000001","\\x00020200"}
 (1 row)
 
-SELECT * FROM fsm_page_contents(get_raw_page('test1', 'fsm', 0));
- fsm_page_contents 
--------------------
- 0: 254           +
- 1: 254           +
- 3: 254           +
- 7: 254           +
- 15: 254          +
- 31: 254          +
- 63: 254          +
- 127: 254         +
- 255: 254         +
- 511: 254         +
- 1023: 254        +
- 2047: 254        +
- 4095: 254        +
- fp_next_slot: 0  +
- 
-(1 row)
-
 DROP TABLE test1;
 -- check that using any of these functions with a partitioned table or index
 -- would fail