This repository has been archived by the owner on May 3, 2024. It is now read-only.
Try less memory usage #516
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
MBkkt
commented
Jul 21, 2023
| @@ -82,45 +82,46 @@ class column final : public irs::column_output { | |||
|
|
|||
| class address_table { | |||
| public: | |||
| address_table(ManagedTypedAllocator<uint64_t> alloc) : alloc_{alloc} { | |||
| offsets_ = alloc_.allocate(kBlockSize); | |||
There was a problem hiding this comment.
Now we allocate very good aligned blocks, so I think it's better utilize allocator.
Also if allocator decide to use mmap, we will get lazy init
There was a problem hiding this comment.
Here would be 2 allocations. One for deque itself and this one for internal contents. Better to benchmark this on test load with many columns.
MBkkt
commented
Jul 21, 2023
| IRS_ASSERT(offset_ > offsets_); | ||
| *--offset_ = 0; | ||
| IRS_ASSERT(offsets_ < offset_); | ||
| --offset_; |
There was a problem hiding this comment.
we don't need to set zero to unused data
MBkkt
commented
Jul 21, 2023
|
|
||
| void reset() noexcept { | ||
| std::memset(offsets_, 0, sizeof offsets_); |
There was a problem hiding this comment.
we don't need to set zero to unused data
MBkkt
commented
Jul 21, 2023
|
|
||
| private: | ||
| uint64_t offsets_[kBlockSize]{}; |
There was a problem hiding this comment.
We don't need zero unused data
MBkkt
commented
Jul 21, 2023
| offset_ = offsets_; | ||
| } | ||
|
|
||
| ~address_table() { alloc_.deallocate(offsets_, kBlockSize); } |
There was a problem hiding this comment.
Also deque with columns works better now:
blocks in deque could handle few columns and we don't store this memory forever.
only while write segment
Dronplane
reviewed
Jul 28, 2023
Dronplane
reviewed
Jul 28, 2023
MBkkt
added a commit
that referenced
this pull request
Aug 14, 2023
* Try less memory usage * fixes * Update columnstore2.cpp * Add benchmark * Add benchmark * Add benchmark
MBkkt
added a commit
that referenced
this pull request
Aug 14, 2023
* Try less memory usage * fixes * Update columnstore2.cpp * Add benchmark * Add benchmark * Add benchmark
MBkkt
added a commit
that referenced
this pull request
Aug 14, 2023
* Try less memory usage (#516) * Try less memory usage * fixes * Update columnstore2.cpp * Add benchmark * Add benchmark * Add benchmark * Make reserve for wildcard nfa construction (#536) * Fix burst trie UB (#537) * Fix UB * Hotfix for directory resource managment * Try less memory usage (#516) * Try less memory usage * fixes * Update columnstore2.cpp * Add benchmark * Add benchmark * Add benchmark * wip (#543) * wip * wip * wip
MBkkt
added a commit
that referenced
this pull request
Aug 14, 2023
* Try less memory usage (#516) * Try less memory usage * fixes * Update columnstore2.cpp * Add benchmark * Add benchmark * Add benchmark * Make reserve for wildcard nfa construction (#536) * Fix burst trie UB (#537) * Fix UB * Hotfix for directory resource managment * Try less memory usage (#516) * Try less memory usage * fixes * Update columnstore2.cpp * Add benchmark * Add benchmark * Add benchmark * wip (#543) * wip * wip * wip
Merged
Merged
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
0.4 vs 0.5 on my machine with old memset.
And 0.1 without unnecessary memset