Reconstruct heap graph from disk graph #519
Closed
Conversation
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
sam-herman
requested review from
MarkWolters,
jshook,
marianotepper and
tlwillke
as code owners
2 tasks
sam-herman
force-pushed
the
reconstruct-heap-graph-from-disk-graph
branch
from
e30816a to
349bd07
Compare
get incremental graph generation working add serialization for NeighborsCache add ord-mapping logic from PQ vectors to RAVV add ord mapping from RAVV to graph creation Signed-off-by: Samuel Herman <sherman8915@gmail.com>
Signed-off-by: Samuel Herman <sherman8915@gmail.com>
sam-herman
force-pushed
the
reconstruct-heap-graph-from-disk-graph
branch
from
7ba0a3a to
279b5aa
Compare
Signed-off-by: Samuel Herman <sherman8915@gmail.com>
Signed-off-by: Samuel Herman <sherman8915@gmail.com>
Signed-off-by: Samuel Herman <sherman8915@gmail.com>
Collaborator
|
Superseded by #536 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
Description
Multiple database projects such as C* and OpenSearch/Solr/Lucene use LSM mechanism with frequent merges.
This in turn creates a high overhead that forces us to reconstruct the entire graph from scratch upon every merge.
A more economic approach would be to pick a leading graph that was previously persisted to disk and incrementally add small graph nodes to it.
This PR makes some of the required changes to support that behavior in upstream systems such as C* and OpenSearch.
Changes
OnDiskGraphIndexfor a faster re-creation of theOnHeapGraphIndexwhen read back from disk. The cache is separate and is optional, therefore we can choose whether to apply it or not, without any breaking changes to the currentOnDiskGraphIndex. This can later be augmented to a format if we choose to.Testing
Added tests for graph overlap and recall for reconstructed graphs.