Merged
Conversation
Instead of caching full graph objects (user and group), let's just cache the data we use. This brings down memory consumption by roughly ~60% on a test machine where there are a lot of objects to cache.
Let's change it so we will only load the resolver cache when it is actually used. If a customer doesn't use PR functionality at all, it would never light up, so we save the bandwidth, CPU and memory. Compared to 0.1.8, this change yields ~80% memory savings on a machine with lots of graph objects.
ivanv-microsoft
approved these changes
Jan 12, 2024
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This is a regression from the PR functionality (#91). We have a cache of graph objects (user and groups) to provide email resolution in query UX. For customers with very large graphs, lo and behold, we tie up a lot of memory.
We are going to mitigate the memory usage with two approaches:
(1) Instead of caching full graph objects (user and group), let's just cache the data we use. This brings down memory consumption by ~60% on a test machine where there are a lot of objects to cache.
(2) We will only load the resolver cache when it is used. If a customer doesn't use PR functionality at all, it would never light up, so we save the bandwidth, CPU, and memory. Compared to 0.1.8, this change yields ~80% memory savings on a machine with lots of graph objects.
Overall, this will make the PR feature fully pay-for-play as you won't pay any CPU, network, or memory costs unless you use it. And even when you use it, we will use a lot less memory (~60% less in my tests).
Also tested with PR queries and observed the on-demand querying happening. All async and not blocking any UI threads as it was previously.
Fixes #100.