Update chaindb to use a slab allocator for small size buffers.#1070
Open
greg7mdp wants to merge 21 commits intorelease/2.0from
Open
Update chaindb to use a slab allocator for small size buffers.#1070greg7mdp wants to merge 21 commits intorelease/2.0from
chaindb to use a slab allocator for small size buffers.#1070greg7mdp wants to merge 21 commits intorelease/2.0from
Conversation
Contributor
|
Note:start |
heifner
reviewed
Dec 16, 2024
Closed
greg7mdp
changed the title
chaindb to use a slab allocator for small size buffers.
heifner
approved these changes
Apr 29, 2025
heifner
reviewed
Nov 17, 2025
| _file_mapping = bip::file_mapping(_data_file_path.generic_string().c_str(), bip::read_write); | ||
| _file_mapped_region = bip::mapped_region(_file_mapping, bip::read_write); | ||
| file_mapped_segment_manager = new ((char*)_file_mapped_region.get_address()+header_size) segment_manager(shared_file_size-header_size); | ||
| new (_file_mapped_region.get_address()) db_header; |
Contributor
There was a problem hiding this comment.
This line must remain when merged.
The change from: 5153e29#diff-4598b34f65fa412d3263ad3b6d1d1711c29a7c1b7689ed659c63a5ab1cc0b682R141
Doesn't work with the small allocator.
heifner
reviewed
Nov 18, 2025
| if(hs.fail()) | ||
| BOOST_THROW_EXCEPTION(std::system_error(make_error_code(db_error_code::bad_header))); | ||
|
|
||
| db_header* dbheader = reinterpret_cast<db_header*>(header); |
Contributor
There was a problem hiding this comment.
This also must remain when merged.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Resolves #1049.
Now this integrates the changes from AntelopeIO/chainbase#54, which can be referred to for previous comments.
As the issue describes, the current chainbase does a lot of small buffer allocations directly through the bip allocator which hits the segment manager directly. Not only this is inefficient (the rbtree of free buffers need to be rebalanced for every alloc/free), it also has a higher memory overhead.
The slab allocators for small buffers (in the updated chainbase submodule) batch allocations of small buffers into larger blocks, reducing the fragmentattion within the shared memory segment, and providing better efficiency and a smaller memory footprint.
For comparison, here are some results comparing snapshot loading on my linux box (release build)
snapshot from Aug 11, 2024
snapshot from Sept 29, 2024
I also expect a slight performance increase for processing transactions, as the
_undo_stackdeque allocations used to go through the segment allocator as well (for blocks of 16undo_state), and these allocations are now batched.