Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Improve cache for MetricLookup.linkable_elements_for_measure() #1443

Merged
merged 1 commit into from
Oct 7, 2024
Merged

Improve cache for MetricLookup.linkable_elements_for_measure() #1443

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 28 additions & 9 deletions metricflow-semantics/metricflow_semantics/model/semantics/metric_lookup.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
import functools
import logging
import time
from typing import Dict, Optional, Sequence, Set
from typing import Dict, Optional, Sequence, Set, Tuple

from dbt_semantic_interfaces.enum_extension import assert_values_exhausted
from dbt_semantic_interfaces.protocols.metric import Metric, MetricInputMeasure, MetricType
Expand All @@ -13,6 +13,7 @@

from metricflow_semantics.errors.error_classes import DuplicateMetricError, MetricNotFoundError, NonExistentMeasureError
from metricflow_semantics.mf_logging.lazy_formattable import LazyFormat
from metricflow_semantics.model.linkable_element_property import LinkableElementProperty
from metricflow_semantics.model.semantics.element_filter import LinkableElementFilter
from metricflow_semantics.model.semantics.linkable_element_set import LinkableElementSet
from metricflow_semantics.model.semantics.linkable_spec_resolver import (
Expand Down Expand Up @@ -62,25 +63,43 @@ def __init__(
MetricReference, Sequence[TimeDimensionSpec]
] = {}

@functools.lru_cache
# Cache for `linkable_elements_for_measure()`.
self._linkable_element_set_for_measure_cache: Dict[
Tuple[MeasureReference, LinkableElementFilter], LinkableElementSet
] = {}

def linkable_elements_for_measure(
self,
measure_reference: MeasureReference,
element_filter: LinkableElementFilter = LinkableElementFilter(),
) -> LinkableElementSet:
"""Return the set of linkable elements reachable from a given measure."""
start_time = time.time()
linkable_element_set = self._linkable_spec_resolver.get_linkable_element_set_for_measure(
measure_reference=measure_reference,
element_filter=element_filter,
)
# Don't cache the result when group-by-metrics are selected as there can be many of them and may significantly
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we remove the cache for group by metrics, doesn't that mean this will be super slow when we try to fetch group by metric options?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I had some concerns about memory usage, so caching of group-by-metrics is handled later through a separate cache that's LRU and avoids GC issues.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Got it 👍

# increase memory usage.
if (
LinkableElementProperty.METRIC in element_filter.with_any_of
and LinkableElementProperty.METRIC not in element_filter.without_any_of
):
return self._linkable_spec_resolver.get_linkable_element_set_for_measure(measure_reference, element_filter)

cache_key = (measure_reference, element_filter)
result = self._linkable_element_set_for_measure_cache.get(cache_key)
if result is not None:
return result

result = self._linkable_spec_resolver.get_linkable_element_set_for_measure(measure_reference, element_filter)
self._linkable_element_set_for_measure_cache[cache_key] = result

logger.debug(
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Curious why you removed this log? Seems helpful to see if there was improvement on performance for this method. Also seems like it would be a good one to make INFO instead of DEBUG so that we can see how it performs for customers with larger manifests.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This method can potentially be called a lot during query parsing, so it's easier to assess the runtime through DD tooling. e.g. a query with ~100 metrics, ~100 group-by items (including ones in filters) leads to a lot of log lines. Could leave a debug line in there though.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added that debug line.

LazyFormat(
lambda: f"Getting valid linkable elements for measure '{measure_reference.element_name}' took: {time.time() - start_time:.2f}s"
"Finished getting linkable elements",
measure_reference=measure_reference,
element_filter=element_filter,
runtime=time.time() - start_time,
)
)

return linkable_element_set
return result

@functools.lru_cache
def linkable_elements_for_no_metrics_query(
Expand Down
Loading