Add optional S3 streaming for cache nodes #1079
Merged
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Outdated suggestions cannot be applied.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.
Summary
Adds an experimental flag for the cache nodes, that optionally streams in the data from object storage. This removes the requirement for disk for cache nodes, enabling switching to significantly cheaper storage types.
Current implementation still results in the assignments still being "downloaded" and "initialized," but transparent to Lucene this is fetching with block-level caches into memory at the time of request.
To use set:
optionally: