Optimized read footer checksum from FileInfo #69415
Merged
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
tlrx
added
>bug
:Distributed Coordination/Snapshot/Restore
elasticmachine
added
the
Team:Distributed (Obsolete)
Collaborator
|
Pinging @elastic/es-distributed (Team:Distributed) |
original-brownbear
approved these changes
Feb 23, 2021
Contributor
original-brownbear
left a comment
original-brownbear
left a comment
There was a problem hiding this comment.
LGTM nice find!
Member
Author
|
Thanks Armin and Yannick |
tlrx
added a commit
to tlrx/elasticsearch
that referenced
this pull request
Feb 23, 2021
Searchable snapshots IndexInput implementations detect read operations that are executed on the last 16 bytes of files which contain the footer checksum, and then serves them from memory (FileInfo) instead of relying on cache mechanisms. But the current implementation expect an exact checksum read: ie a read operation that starts at file length - 16 bytes and expects to read exactly 16 bytes. This is not always the case as IndexInput implementations extend BufferedIndexInput which uses an internal buffer of 1024 bytes: in some case the remaining length (ie, the last portion of the file) might be less than 16 bytes long and thus a very small read is engaged using cache (or direct) read where it could instead be also served from memory.
tlrx
added a commit
to tlrx/elasticsearch
that referenced
this pull request
Feb 23, 2021
Searchable snapshots IndexInput implementations detect read operations that are executed on the last 16 bytes of files which contain the footer checksum, and then serves them from memory (FileInfo) instead of relying on cache mechanisms. But the current implementation expect an exact checksum read: ie a read operation that starts at file length - 16 bytes and expects to read exactly 16 bytes. This is not always the case as IndexInput implementations extend BufferedIndexInput which uses an internal buffer of 1024 bytes: in some case the remaining length (ie, the last portion of the file) might be less than 16 bytes long and thus a very small read is engaged using cache (or direct) read where it could instead be also served from memory.
tlrx
added a commit
that referenced
this pull request
Feb 23, 2021
Searchable snapshots IndexInput implementations detect read operations that are executed on the last 16 bytes of files which contain the footer checksum, and then serves them from memory (FileInfo) instead of relying on cache mechanisms. But the current implementation expect an exact checksum read: ie a read operation that starts at file length - 16 bytes and expects to read exactly 16 bytes. This is not always the case as IndexInput implementations extend BufferedIndexInput which uses an internal buffer of 1024 bytes: in some case the remaining length (ie, the last portion of the file) might be less than 16 bytes long and thus a very small read is engaged using cache (or direct) read where it could instead be also served from memory. Backport of #69415
tlrx
added a commit
that referenced
this pull request
Feb 23, 2021
Searchable snapshots IndexInput implementations detect read operations that are executed on the last 16 bytes of files which contain the footer checksum, and then serves them from memory (FileInfo) instead of relying on cache mechanisms. But the current implementation expect an exact checksum read: ie a read operation that starts at file length - 16 bytes and expects to read exactly 16 bytes. This is not always the case as IndexInput implementations extend BufferedIndexInput which uses an internal buffer of 1024 bytes: in some case the remaining length (ie, the last portion of the file) might be less than 16 bytes long and thus a very small read is engaged using cache (or direct) read where it could instead be also served from memory. Backport of #69415
This was referenced Feb 23, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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.
Searchable snapshots
IndexInputimplementations detect read operations that are executed on the last16bytes of files which contain the footer checksum, and then serves them from memory (FileInfo) instead of relying on cache mechanisms.But the current implementation expect an exact checksum read: ie a read operation that starts at file
length - 16 bytesand expects to read exactly16bytes. This is not always the case asIndexInputimplementations extendBufferedIndexInputwhich uses an internal buffer of1024bytes: in some case the remaining length (ie, the last portion of the file) might be less than 16 bytes long and thus a very small read is engaged using cache (or direct) read where it could instead be also served from memory.