mirrors
/
postgres
mirror of https://github.com/postgres/postgres
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Improve performance of dumpSequenceData(). 

As one might guess, this function dumps the sequence data.  It is
called once per sequence, and each such call executes a query to
retrieve the relevant data for a single sequence.  This can cause
pg_dump to take significantly longer, especially when there are
many sequences.

This commit improves the performance of this function by gathering
all the sequence data with a single query at the beginning of
pg_dump.  This information is stored in a sorted array that
dumpSequenceData() can bsearch() for what it needs.  This follows a
similar approach as previous commits that introduced sorted arrays
for role information, pg_class information, and sequence metadata.
As with those commits, this patch will cause pg_dump to use more
memory, but that isn't expected to be too egregious.

Note that we use the brand new function pg_sequence_read_tuple() in
the query that gathers all sequence data, so we must continue to
use the preexisting query-per-sequence approach for versions older
than 18.

Reviewed-by: Euler Taveira, Michael Paquier, Tom Lane
Discussion: https://postgr.es/m/20240503025140.GA1227404%40nathanxps13
This commit is contained in:
Nathan Bossart 2024-07-31 10:12:42 -05:00
parent c8b06bb969
commit bd15b7db48
1 changed files with 64 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

63
src/bin/pg_dump/pg_dump.c
View File

@ -132,6 +132,8 @@ typedef struct
int64 startv; /* start value */
int64 incby; /* increment value */
int64 cache; /* cache size */
int64 last_value; /* last value of sequence */
bool is_called; /* whether nextval advances before returning */
} SequenceItem;
typedef enum OidOptions
@ -17330,16 +17332,30 @@ collectSequences(Archive *fout)
* Before Postgres 10, sequence metadata is in the sequence itself. With
* some extra effort, we might be able to use the sorted table for those
* versions, but for now it seems unlikely to be worth it.
*
* Since version 18, we can gather the sequence data in this query with
* pg_sequence_read_tuple(), but we only do so for non-schema-only dumps.
*/
if (fout->remoteVersion < 100000)
return;
else if (fout->remoteVersion < 180000 ||
(fout->dopt->schemaOnly && !fout->dopt->sequence_data))
query = "SELECT seqrelid, format_type(seqtypid, NULL), "
"seqstart, seqincrement, "
"seqmax, seqmin, "
"seqcache, seqcycle, "
"NULL, 'f' "
"FROM pg_catalog.pg_sequence "
"ORDER BY seqrelid";
else
query = "SELECT seqrelid, format_type(seqtypid, NULL), "
"seqstart, seqincrement, "
"seqmax, seqmin, "
"seqcache, seqcycle "
"FROM pg_catalog.pg_sequence "
"ORDER BY seqrelid";
"seqcache, seqcycle, "
"last_value, is_called "
"FROM pg_catalog.pg_sequence, "
"pg_sequence_read_tuple(seqrelid) "
"ORDER BY seqrelid;";
res = ExecuteSqlQuery(fout, query, PGRES_TUPLES_OK);
@ -17356,6 +17372,8 @@ collectSequences(Archive *fout)
sequences[i].minv = strtoi64(PQgetvalue(res, i, 5), NULL, 10);
sequences[i].cache = strtoi64(PQgetvalue(res, i, 6), NULL, 10);
sequences[i].cycled = (strcmp(PQgetvalue(res, i, 7), "t") == 0);
sequences[i].last_value = strtoi64(PQgetvalue(res, i, 8), NULL, 10);
sequences[i].is_called = (strcmp(PQgetvalue(res, i, 9), "t") == 0);
}
PQclear(res);
@ -17622,11 +17640,22 @@ static void
dumpSequenceData(Archive *fout, const TableDataInfo *tdinfo)
{
TableInfo *tbinfo = tdinfo->tdtable;
PGresult *res;
char *last;
int64 last;
bool called;
PQExpBuffer query = createPQExpBuffer();
/*
* For versions >= 18, the sequence information is gathered in the sorted
* array before any calls to dumpSequenceData(). See collectSequences()
* for more information.
*
* For older versions, we have to query the sequence relations
* individually.
*/
if (fout->remoteVersion < 180000)
{
PGresult *res;
appendPQExpBuffer(query,
"SELECT last_value, is_called FROM %s",
fmtQualifiedDumpable(tbinfo));
@ -17639,13 +17668,31 @@ dumpSequenceData(Archive *fout, const TableDataInfo *tdinfo)
PQntuples(res)),
tbinfo->dobj.name, PQntuples(res));
last = PQgetvalue(res, 0, 0);
last = strtoi64(PQgetvalue(res, 0, 0), NULL, 10);
called = (strcmp(PQgetvalue(res, 0, 1), "t") == 0);
PQclear(res);
}
else
{
SequenceItem key = {0};
SequenceItem *entry;
Assert(sequences);
Assert(tbinfo->dobj.catId.oid);
key.oid = tbinfo->dobj.catId.oid;
entry = bsearch(&key, sequences, nsequences,
sizeof(SequenceItem), SequenceItemCmp);
last = entry->last_value;
called = entry->is_called;
}
resetPQExpBuffer(query);
appendPQExpBufferStr(query, "SELECT pg_catalog.setval(");
appendStringLiteralAH(query, fmtQualifiedDumpable(tbinfo), fout);
appendPQExpBuffer(query, ", %s, %s);\n",
appendPQExpBuffer(query, ", " INT64_FORMAT ", %s);\n",
last, (called ? "true" : "false"));
if (tdinfo->dobj.dump & DUMP_COMPONENT_DATA)
@ -17659,8 +17706,6 @@ dumpSequenceData(Archive *fout, const TableDataInfo *tdinfo)
.deps = &(tbinfo->dobj.dumpId),
.nDeps = 1));
PQclear(res);
destroyPQExpBuffer(query);
}