HBASE-27329 Introduce prefix tree index block encoding use less space #4782
+2,667
−3
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
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
🎊 +1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
Apache9
reviewed
Sep 18, 2022
| throw new IllegalStateException("Unexpected unable to find index=" + index); | ||
| } | ||
|
|
||
| public static class TokenizerNode { |
There was a problem hiding this comment.
I think the memory overhead is a bit large here? Is it possible to use a double array trie here?
There was a problem hiding this comment.
This is only need when serialize, and will not use it when reading.
There was a problem hiding this comment.
Then what is the IndexEncodedSeeker used for?
There was a problem hiding this comment.
IndexEncodedSeeker is used for reading, and read ByteBuffer data direct.
There was a problem hiding this comment.
OK, so this TokenizerNode is just for construct the trie and serialize it, when reading, we will use the compact format and read the ByteBuffer directly instead of deserialize it to a TokenizerNode? Please add more comments so others could know this when they are not familiar with the code.
|
The first version is simple, index block only contains cell's row, so one row's all cells need to store in one HFileBlock. The version 2 support row + qualifier + timestamp + type which keep consistent with the current situation, one row's all cells can store in multiple HFileBlock. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
| import org.apache.yetus.audience.InterfaceAudience; | ||
|
|
||
| @InterfaceAudience.Private | ||
| public class PrefixTreeIndexBlockEncoderV2 implements IndexBlockEncoder { |
There was a problem hiding this comment.
Why are there 2 versions of this encoder? What's the difference? When to use one vs the other?
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.
A simple implementation which only consider cell row and not cell qualifier.
00c7-202206201519-wx0t
00c7-202206201519-wx0zcldi7lnsiyas-N
00c7-202206201520-wx0re
00c7-202206201520-wx0ulgrwi7d542tm-N
00c7-202206201520-wx0x7
00c7-202206201521
00c7-202206201521-wx05xfbtw2mopyhs-C
00c7-202206201521-wx08
00c7-202206201521-wx0c
00c7-202206201521-wx0go
00c7-202206201522-wx0t
00c8-202206200751-wx0ah4gnbwptdyna-F
The prefix tree node is like this:
