Optimise async get event lookups

changelog.d/13435.misc

@@ -1 +1 @@
-Prevent unnecessary lookups to any external get event cache. Contributed by Nick @ Beeper (@fizzadar).
+Prevent unnecessary lookups to any external `get_event` cache. Contributed by Nick @ Beeper (@fizzadar).

synapse/storage/databases/main/events_worker.py

@@ -602,7 +602,8 @@ async def _get_events_from_cache_or_db(
  """
  # Shortcut: check if we have any events in the *in memory* cache - this function
  # may be called repeatedly for the same event so at this point we cannot reach
- # out to any external cache for performance reasons.
+ # out to any external cache for performance reasons, this is done later on in
+ # the `get_missing_events_from_cache_or_db` function below.
  event_entry_map = self._get_events_from_local_cache(
  event_ids,
  )
@@ -664,14 +665,13 @@ async def get_missing_events_from_cache_or_db() -> Dict[
  #
  missing_events = {}
  try:
- # First fetch from the cache - including any external caches
- cache_missing_events = await self._get_events_from_cache(
+ # Try to fetch from any external cache, in-memory cache checked above
+ missing_events = await self._get_events_from_external_cache(
  missing_events_ids,
  )
- missing_events.update(cache_missing_events)
  # Now actually fetch any remaining events from the DB
  db_missing_events = await self._get_events_from_db(
- missing_events_ids - set(cache_missing_events.keys()),
+ missing_events_ids - missing_events.keys(),
  )
  missing_events.update(db_missing_events)
  except Exception as e:
@@ -780,9 +780,30 @@ async def _get_events_from_cache(
  )
 
  missing_event_ids = {e for e in events if e not in event_map}
+ event_map.update(
+ await self._get_events_from_external_cache(
+ events=missing_event_ids,
+ update_metrics=update_metrics,
+ )
+ )
+
+ return event_map
+
+ async def _get_events_from_external_cache(
+ self, events: Iterable[str], update_metrics: bool = True
+ ) -> Dict[str, EventCacheEntry]:
+ """Fetch events from any configured external cache.
+
+ May return rejected events.
+
+ Args:
+ events: list of event_ids to fetch
+ update_metrics: Whether to update the cache hit ratio metrics
+ """
+ event_map = {}
 
- for event_id in missing_event_ids:
- ret = await self._get_event_cache.get(
+ for event_id in events:
+ ret = await self._get_event_cache.get_external(
  (event_id,), None, update_metrics=update_metrics
  )
  if ret:

synapse/util/caches/lrucache.py

@@ -834,6 +834,15 @@ async def get(
  ) -> Optional[VT]:
  return self._lru_cache.get(key, update_metrics=update_metrics)
 
+ async def get_external(
+ self,
+ key: KT,
+ default: Optional[T] = None,
+ update_metrics: bool = True,
+ ) -> Optional[VT]:
+ # This method should fetch from any configured external cache, in this case noop.
+ return None
+
  def get_local(
  self, key: KT, default: Optional[T] = None, update_metrics: bool = True
  ) -> Optional[VT]: