From ba5f73283303e0611c9e597ab6b22a07e4400596 Mon Sep 17 00:00:00 2001 From: Kyle M Hall Date: Mon, 9 Aug 2021 13:46:00 -0400 Subject: [PATCH] Bug 28833: Speed up holds queue builder via parallel processing The holds queue builder can take a very long time to run on systems with many holds. For example, a partner with 124,784 unfilled ( not found ) holds, is taking about 64 minutes to run. If we run that same number of holds in 5 parallel chunks ( splitting the number of records as evenly as possible, but *not* taking into account the holds per bib ), it takes 21.5 minutes. If we use 10 loops, it takes less then 14 minutes. Test Plan: 0) Install the Perl library Parallel::ForkManager 1) Generate a huge number of holds ( a few thousand at the minimum ) 2) Run the holds queue builder, use the `time` utility to track how much time it took to run 3) Set HoldsQueueParallelLoopsCount to 10 4) Repeat step 2, note the improvement in speed 5) Experiment with other values for HoldsQueueParallelLoopsCount 6) prove t/db_dependent/HoldsQueue.t --- C4/HoldsQueue.pm | 50 ++++++++++++++++++------ debian/templates/koha-conf-site.xml.in | 4 ++ etc/koha-conf.xml | 4 ++ misc/cronjobs/holds/build_holds_queue.pl | 4 +- 4 files changed, 50 insertions(+), 12 deletions(-) diff --git a/C4/HoldsQueue.pm b/C4/HoldsQueue.pm index 06b0b2a2ef..6b865c2363 100644 --- a/C4/HoldsQueue.pm +++ b/C4/HoldsQueue.pm @@ -30,8 +30,10 @@ use Koha::Items; use Koha::Patrons; use Koha::Libraries; -use List::Util qw( shuffle ); use List::MoreUtils qw( any ); +use List::Util qw( shuffle ); +use POSIX qw(ceil); +use Parallel::ForkManager; our (@ISA, @EXPORT_OK); BEGIN { @@ -173,16 +175,13 @@ Top level function that turns reserves into tmp_holdsqueue and hold_fill_targets =cut sub CreateQueue { - my $dbh = C4::Context->dbh; + my $loops = shift || 1; + + my $dbh = C4::Context->dbh; $dbh->do("DELETE FROM tmp_holdsqueue"); # clear the old table for new info $dbh->do("DELETE FROM hold_fill_targets"); - my $total_bibs = 0; - my $total_requests = 0; - my $total_available_items = 0; - my $num_items_mapped = 0; - my $branches_to_use; my $transport_cost_matrix; my $use_transport_cost_matrix = C4::Context->preference("UseTransportCostMatrix"); @@ -199,19 +198,48 @@ sub CreateQueue { my $bibs_with_pending_requests = GetBibsWithPendingHoldRequests(); + # Split the list of bibs into groups to run in parallel + if ( $loops > 1 ) { + my $bibs_per_chunk = ceil( scalar @$bibs_with_pending_requests / $loops ); + $bibs_per_chunk = @$bibs_with_pending_requests if @$bibs_with_pending_requests <= $bibs_per_chunk; + my @chunks; + + push( @chunks, [ splice @$bibs_with_pending_requests, 0, $bibs_per_chunk ] ) while @$bibs_with_pending_requests; + push( @{$chunks[0]}, @$bibs_with_pending_requests ); # Add any remainders to the first parallel process + + my $pm = Parallel::ForkManager->new($loops); + + DATA_LOOP: + foreach my $chunk (@chunks) { + my $pid = $pm->start and next DATA_LOOP; + _ProcessBiblios( $chunk, $branches_to_use, $transport_cost_matrix ); + $pm->finish; + } + + $pm->wait_all_children; + } else { + _ProcessBiblios( $bibs_with_pending_requests, $branches_to_use, $transport_cost_matrix ); + } +} + +=head2 _ProcessBiblios + +=cut + +sub _ProcessBiblios { + my $bibs_with_pending_requests = shift; + my $branches_to_use = shift; + my $transport_cost_matrix = shift; + foreach my $biblionumber (@$bibs_with_pending_requests) { - $total_bibs++; my $hold_requests = GetPendingHoldRequestsForBib($biblionumber); my $available_items = GetItemsAvailableToFillHoldRequestsForBib($biblionumber, $branches_to_use); - $total_requests += scalar(@$hold_requests); - $total_available_items += scalar(@$available_items); my $item_map = MapItemsToHoldRequests($hold_requests, $available_items, $branches_to_use, $transport_cost_matrix); $item_map or next; my $item_map_size = scalar(keys %$item_map) or next; - $num_items_mapped += $item_map_size; CreatePicklistFromItemMap($item_map); AddToHoldTargetMap($item_map); if (($item_map_size < scalar(@$hold_requests )) and diff --git a/debian/templates/koha-conf-site.xml.in b/debian/templates/koha-conf-site.xml.in index 6525e4ce97..e27a9806dd 100644 --- a/debian/templates/koha-conf-site.xml.in +++ b/debian/templates/koha-conf-site.xml.in @@ -449,5 +449,9 @@ __END_SRU_PUBLICSERVER__ __MESSAGE_BROKER_VHOST__ + + 1 + + diff --git a/etc/koha-conf.xml b/etc/koha-conf.xml index 118b6b8207..5e00ea6475 100644 --- a/etc/koha-conf.xml +++ b/etc/koha-conf.xml @@ -266,5 +266,9 @@ + + 1 + + diff --git a/misc/cronjobs/holds/build_holds_queue.pl b/misc/cronjobs/holds/build_holds_queue.pl index 6e1437dcd3..2b9fff4965 100755 --- a/misc/cronjobs/holds/build_holds_queue.pl +++ b/misc/cronjobs/holds/build_holds_queue.pl @@ -18,8 +18,10 @@ BEGIN { use Koha::Script -cron; use C4::HoldsQueue qw(CreateQueue); use C4::Log qw( cronlogaction ); +use C4::Context; cronlogaction(); -CreateQueue(); +my $loops = C4::Context->config('holds_queue_builder')->{parallel_loops_count}; +CreateQueue($loops); -- 2.30.1 (Apple Git-130)