From: Sander Saarend Date: Mon, 17 Jul 2023 09:05:55 +0000 (+0300) Subject: Optimize hot rank updates (#3617) X-Git-Url: http://these/git/%7B%60%24%7BwebArchiveUrl%7D/save/static/git-logo.png?a=commitdiff_plain;h=6688a8a5d44397d9dde5e0e082ef4e9a8cb5ad9e;p=lemmy.git Optimize hot rank updates (#3617) --- diff --git a/migrations/2023-07-14-215339_aggregates_nonzero_indexes/down.sql b/migrations/2023-07-14-215339_aggregates_nonzero_indexes/down.sql new file mode 100644 index 00000000..3e247b58 --- /dev/null +++ b/migrations/2023-07-14-215339_aggregates_nonzero_indexes/down.sql @@ -0,0 +1,4 @@ +-- This file should undo anything in `up.sql` +DROP INDEX idx_community_aggregates_nonzero_hotrank; +DROP INDEX idx_comment_aggregates_nonzero_hotrank; +DROP INDEX idx_post_aggregates_nonzero_hotrank; \ No newline at end of file diff --git a/migrations/2023-07-14-215339_aggregates_nonzero_indexes/up.sql b/migrations/2023-07-14-215339_aggregates_nonzero_indexes/up.sql new file mode 100644 index 00000000..2d3cd3b2 --- /dev/null +++ b/migrations/2023-07-14-215339_aggregates_nonzero_indexes/up.sql @@ -0,0 +1,4 @@ +-- Your SQL goes here +CREATE INDEX idx_community_aggregates_nonzero_hotrank ON community_aggregates (published) WHERE hot_rank != 0; +CREATE INDEX idx_comment_aggregates_nonzero_hotrank ON comment_aggregates (published) WHERE hot_rank != 0; +CREATE INDEX idx_post_aggregates_nonzero_hotrank ON post_aggregates (published DESC) WHERE hot_rank != 0 OR hot_rank_active != 0; \ No newline at end of file diff --git a/src/scheduled_tasks.rs b/src/scheduled_tasks.rs index ad97d193..a4d09268 100644 --- a/src/scheduled_tasks.rs +++ b/src/scheduled_tasks.rs @@ -58,7 +58,7 @@ pub fn setup( let url = db_url.clone(); scheduler.every(CTimeUnits::minutes(15)).run(move || { let mut conn = PgConnection::establish(&url).expect("could not establish connection"); - update_hot_ranks(&mut conn, true); + update_hot_ranks(&mut conn); }); // Delete any captcha answers older than ten minutes, every ten minutes @@ -107,7 +107,7 @@ pub fn setup( fn startup_jobs(db_url: &str) { let mut conn = PgConnection::establish(db_url).expect("could not establish connection"); active_counts(&mut conn); - update_hot_ranks(&mut conn, false); + update_hot_ranks(&mut conn); update_banned_when_expired(&mut conn); clear_old_activities(&mut conn); overwrite_deleted_posts_and_comments(&mut conn); @@ -115,35 +115,29 @@ fn startup_jobs(db_url: &str) { /// Update the hot_rank columns for the aggregates tables /// Runs in batches until all necessary rows are updated once -fn update_hot_ranks(conn: &mut PgConnection, last_week_only: bool) { - let process_start_time = if last_week_only { - info!("Updating hot ranks for last week..."); - naive_now() - chrono::Duration::days(7) - } else { - info!("Updating hot ranks for all history..."); - NaiveDateTime::from_timestamp_opt(0, 0).expect("0 timestamp creation") - }; +fn update_hot_ranks(conn: &mut PgConnection) { + info!("Updating hot ranks for all history..."); process_hot_ranks_in_batches( conn, "post_aggregates", + "a.hot_rank != 0 OR a.hot_rank_active != 0", "SET hot_rank = hot_rank(a.score, a.published), hot_rank_active = hot_rank(a.score, a.newest_comment_time_necro)", - process_start_time, ); process_hot_ranks_in_batches( conn, "comment_aggregates", + "a.hot_rank != 0", "SET hot_rank = hot_rank(a.score, a.published)", - process_start_time, ); process_hot_ranks_in_batches( conn, "community_aggregates", + "a.hot_rank != 0", "SET hot_rank = hot_rank(a.subscribers, a.published)", - process_start_time, ); info!("Finished hot ranks update!"); @@ -155,18 +149,20 @@ struct HotRanksUpdateResult { published: NaiveDateTime, } -/// Runs the hot rank update query in batches until all rows after `process_start_time` have been -/// processed. -/// In `set_clause`, "a" will refer to the current aggregates table. +/// Runs the hot rank update query in batches until all rows have been processed. +/// In `where_clause` and `set_clause`, "a" will refer to the current aggregates table. /// Locked rows are skipped in order to prevent deadlocks (they will likely get updated on the next /// run) fn process_hot_ranks_in_batches( conn: &mut PgConnection, table_name: &str, + where_clause: &str, set_clause: &str, - process_start_time: NaiveDateTime, ) { + let process_start_time = NaiveDateTime::from_timestamp_opt(0, 0).expect("0 timestamp creation"); + let update_batch_size = 1000; // Bigger batches than this tend to cause seq scans + let mut processed_rows_count = 0; let mut previous_batch_result = Some(process_start_time); while let Some(previous_batch_last_published) = previous_batch_result { // Raw `sql_query` is used as a performance optimization - Diesel does not support doing this @@ -174,7 +170,7 @@ fn process_hot_ranks_in_batches( let result = sql_query(format!( r#"WITH batch AS (SELECT a.id FROM {aggregates_table} a - WHERE a.published > $1 + WHERE a.published > $1 AND ({where_clause}) ORDER BY a.published LIMIT $2 FOR UPDATE SKIP LOCKED) @@ -182,14 +178,18 @@ fn process_hot_ranks_in_batches( FROM batch WHERE a.id = batch.id RETURNING a.published; "#, aggregates_table = table_name, - set_clause = set_clause + set_clause = set_clause, + where_clause = where_clause )) .bind::(previous_batch_last_published) .bind::(update_batch_size) .get_results::(conn); match result { - Ok(updated_rows) => previous_batch_result = updated_rows.last().map(|row| row.published), + Ok(updated_rows) => { + processed_rows_count += updated_rows.len(); + previous_batch_result = updated_rows.last().map(|row| row.published); + } Err(e) => { error!("Failed to update {} hot_ranks: {}", table_name, e); break; @@ -197,8 +197,8 @@ fn process_hot_ranks_in_batches( } } info!( - "Finished process_hot_ranks_in_batches execution for {}", - table_name + "Finished process_hot_ranks_in_batches execution for {} (processed {} rows)", + table_name, processed_rows_count ); }