Optionally prefetch referenced data in recovery.

Introduce a new GUC recovery_prefetch, disabled by default.  When
enabled, look ahead in the WAL and try to initiate asynchronous reading
of referenced data blocks that are not yet cached in our buffer pool.
For now, this is done with posix_fadvise(), which has several caveats.
Better mechanisms will follow in later work on the I/O subsystem.

The GUC maintenance_io_concurrency is used to limit the number of
concurrent I/Os we allow ourselves to initiate, based on pessimistic
heuristics used to infer that I/Os have begun and completed.

The GUC wal_decode_buffer_size is used to limit the maximum distance we
are prepared to read ahead in the WAL to find uncached blocks.

Reviewed-by: Alvaro Herrera <alvherre@2ndquadrant.com> (parts)
Reviewed-by: Andres Freund <andres@anarazel.de> (parts)
Reviewed-by: Tomas Vondra <tomas.vondra@2ndquadrant.com> (parts)
Tested-by: Tomas Vondra <tomas.vondra@2ndquadrant.com>
Tested-by: Jakub Wartak <Jakub.Wartak@tomtom.com>
Tested-by: Dmitry Dolgov <9erthalion6@gmail.com>
Tested-by: Sait Talha Nisanci <Sait.Nisanci@microsoft.com>
Discussion: https://postgr.es/m/CA%2BhUKGJ4VJN8ttxScUFM8dOKX0BrBiboo5uz1cq%3DAovOddfHpA%40mail.gmail.com

1 files changed, 50 insertions, 1 deletions

diff --git a/src/backend/utils/misc/guc.c b/src/backend/utils/misc/guc.c
index bee976bae87..090abdad8ba 100644
--- a/src/backend/utils/misc/guc.c
+++ b/src/backend/utils/misc/guc.c
@@ -41,6 +41,7 @@
 #include "access/twophase.h"
 #include "access/xact.h"
 #include "access/xlog_internal.h"
+#include "access/xlogprefetch.h"
 #include "catalog/namespace.h"
 #include "catalog/pg_authid.h"
 #include "catalog/storage.h"
@@ -209,6 +210,7 @@ static bool check_effective_io_concurrency(int *newval, void **extra, GucSource
 static bool check_maintenance_io_concurrency(int *newval, void **extra, GucSource source);
 static bool check_huge_page_size(int *newval, void **extra, GucSource source);
 static bool check_client_connection_check_interval(int *newval, void **extra, GucSource source);
+static void assign_maintenance_io_concurrency(int newval, void *extra);
 static void assign_pgstat_temp_directory(const char *newval, void *extra);
 static bool check_application_name(char **newval, void **extra, GucSource source);
 static void assign_application_name(const char *newval, void *extra);
@@ -1294,6 +1296,27 @@ static struct config_bool ConfigureNamesBool[] =
 		true,
 		NULL, NULL, NULL
 	},
+	{
+		{"recovery_prefetch", PGC_SIGHUP, WAL_SETTINGS,
+			gettext_noop("Prefetch referenced blocks during recovery"),
+			gettext_noop("Read ahead of the current replay position to find uncached blocks.")
+		},
+		&recovery_prefetch,
+		false,
+		NULL, assign_recovery_prefetch, NULL
+	},
+	{
+		{"recovery_prefetch_fpw", PGC_SIGHUP, WAL_SETTINGS,
+			gettext_noop("Prefetch blocks that have full page images in the WAL"),
+			gettext_noop("On some systems, there is no benefit to prefetching pages that will be "
+						 "entirely overwritten, but if the logical page size of the filesystem is "
+						 "larger than PostgreSQL's, this can be beneficial.  This option has no "
+						 "effect unless recovery_prefetch is enabled.")
+		},
+		&recovery_prefetch_fpw,
+		false,
+		NULL, assign_recovery_prefetch_fpw, NULL
+	},
 
 	{
 		{"wal_log_hints", PGC_POSTMASTER, WAL_SETTINGS,
@@ -2749,6 +2772,17 @@ static struct config_int ConfigureNamesInt[] =
 	},
 
 	{
+		{"wal_decode_buffer_size", PGC_POSTMASTER, WAL_ARCHIVE_RECOVERY,
+			gettext_noop("Maximum buffer size for reading ahead in the WAL during recovery."),
+			gettext_noop("This controls the maximum distance we can read ahead n the WAL to prefetch referenced blocks."),
+			GUC_UNIT_BYTE
+		},
+		&wal_decode_buffer_size,
+		512 * 1024, 64 * 1024, INT_MAX,
+		NULL, NULL, NULL
+	},
+
+	{
 		{"wal_keep_size", PGC_SIGHUP, REPLICATION_SENDING,
 			gettext_noop("Sets the size of WAL files held for standby servers."),
 			NULL,
@@ -3068,7 +3102,8 @@ static struct config_int ConfigureNamesInt[] =
 		0,
 #endif
 		0, MAX_IO_CONCURRENCY,
-		check_maintenance_io_concurrency, NULL, NULL
+		check_maintenance_io_concurrency, assign_maintenance_io_concurrency,
+		NULL
 	},
 
 	{
@@ -12073,6 +12108,20 @@ check_client_connection_check_interval(int *newval, void **extra, GucSource sour
 }
 
 static void
+assign_maintenance_io_concurrency(int newval, void *extra)
+{
+#ifdef USE_PREFETCH
+	/*
+	 * Reconfigure recovery prefetching, because a setting it depends on
+	 * changed.
+	 */
+	maintenance_io_concurrency = newval;
+	if (AmStartupProcess())
+		XLogPrefetchReconfigure();
+#endif
+}
+
+static void
 assign_pgstat_temp_directory(const char *newval, void *extra)
 {
 	/* check_canonical_path already canonicalized newval for us */