cloud_storage: Rehash segment name prefixes #18762
Merged
Conversation
This file contains 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
abhijat
reviewed
Jun 4, 2024
abhijat
reviewed
Jun 4, 2024
abhijat
reviewed
Jun 4, 2024
abhijat
reviewed
Jun 4, 2024
abhijat
reviewed
Jun 4, 2024
abhijat
previously approved these changes
Jun 4, 2024
micheleRP
reviewed
Jun 4, 2024
src/v/config/configuration.cc
Outdated
| *this, | ||
| "cloud_storage_cache_num_buckets", | ||
| "Divide cloud storage cache across specified number of buckets. This " | ||
| "only works for objects with randomized prefix.", |
There was a problem hiding this comment.
Suggested change
| "only works for objects with randomized prefix.", | |
| "only works for objects with randomized prefixes.", |
The configuration option enables bucketing of the segment and manifest files using different prefixes.
The names that start with randomized prefix are now a subject to rehashing procedure. The prefix is replaced with another prefix which is computed from the hash. This new prefix has lower cardinality. During the lookup we will try to probe two paths. The rehashed and original because the cache might store objects from before this feature was enabled.
Lazin
force-pushed
the
fix/slow-trim-problem
branch
from
0fa4326 to
46609dc
Compare
abhijat
approved these changes
Jun 5, 2024
|
/backport v24.1.x |
|
/backport v23.3.x |
|
Failed to create a backport PR to v23.3.x branch. I tried: |
7 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Currently, the cache uses segment object names as file paths. This means that we have series of nested directories to store every file. Also, the first directory is a randomized prefix. This creates very large number of nonsense directories in the root of the cache.
This PR adds new configuration option
cloud_storage_cache_num_buckets. When this property is set the cache will check if the object name starts with randomized prefix (8-character hex number). If this is the case it will replace it with a different prefix that has much smaller cardinality. Basically, it will put all object names intocloud_storage_cache_num_bucketsbuckets.Backports Required
Release Notes
Features
cloud_storage_cache_num_bucketsconfiguration parameter.