You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@paimon.apache.org by "gripleaf (via GitHub)" <gi...@apache.org> on 2026/08/24 06:11:13 UTC

[I] [Feature] Improve scan performance by reducing shared_ptr contention in hot read paths [paimon-cpp]

gripleaf opened a new issue, #240:
URL: https://github.com/apache/paimon-cpp/issues/240

   ### Search before asking
   
   - [x] I searched in the [issues](https://github.com/apache/paimon-cpp/issues) and found nothing similar.
   
   
   ### Motivation
   
   Recent profiling of highly concurrent scans has revealed several performance bottlenecks caused by
     repeated operations on Arrow-returned `shared_ptr` objects in hot read loops.
   
     Two significant cases have been identified:
   
     1. Manifest readers repeatedly call `StructArray::fields()` while processing individual rows. This
        copies `shared_ptr<Array>` objects and, with GCC 8.3's libstdc++, can introduce substantial lock
        contention through `_Sp_locker`, `pthread_mutex_lock`, and futex waits when multiple workers read
        manifests concurrently.
   
     2. Avro decoding calls `ArrayBuilder::type()` for every integer and timestamp value. Because this
        method returns `std::shared_ptr<DataType>` by value, concurrent scans repeatedly modify reference
        counts on shared Arrow primitive data types, causing cache-line contention. Profiling showed
        `ArrayBuilder::type()` and shared-pointer release operations accounting for a large proportion of
        samples after the manifest bottleneck was removed.
   
     This issue tracks the broader effort to identify and eliminate similar shared-pointer operations
     from scan hot paths. The goal is to cache immutable Arrow metadata at an appropriate batch, reader,
     or builder lifetime, while ensuring caches are invalidated whenever the corresponding Arrow object
     tree is replaced.
   
     The expected outcome is lower synchronization and reference-counting overhead under concurrent
     scans, allowing CPU time to return to actual decoding, memory copying, and buffer management.
   
   ### Solution
   
   _No response_
   
   ### Anything else?
   
   _No response_
   
   ### Are you willing to submit a PR?
   
   - [ ] I'm willing to submit a PR!


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [I] [Feature] Improve scan performance in hot read paths [paimon-cpp]

Posted by "gripleaf (via GitHub)" <gi...@apache.org>.
gripleaf closed issue #240: [Feature] Improve scan performance in hot read paths
URL: https://github.com/apache/paimon-cpp/issues/240


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [I] [Feature] Improve scan performance in hot read paths [paimon-cpp]

Posted by "gripleaf (via GitHub)" <gi...@apache.org>.
gripleaf commented on issue #240:
URL: https://github.com/apache/paimon-cpp/issues/240#issuecomment-5392064310

   A follow-up PR has been opened for this issue: #242 
   
     After profiling the updated implementation, several additional sources of Arrow-side overhead remained.
   
     This follow-up PR focuses on three areas:
   
     1. Avoid redundant Arrow array validation
   
        Arrays produced by controlled Avro builders were validated again after `Finish()`. For nested Struct, List, and Map types, this validation recursively compares data types and computes type fingerprints.
   
        The PR skips this validation in release builds while retaining it in debug builds for invariant checking.
   
     2. Reuse `ColumnarRow` within a manifest batch
   
        The manifest bucket scan previously constructed and destroyed one `ColumnarRow` per row, although the underlying Arrow fields and memory pool remain unchanged throughout the batch.
   
        The PR creates one row view per batch and updates only its row index during iteration.
   
     3. Reserve Avro builder capacity
   
        The target batch size is known before decoding, but Arrow builders previously relied on progressive capacity growth.
   
        The PR reserves capacity for the root builder and same-cardinality Struct children using the configured batch size. List and Map child builders are reserved using their decoded Avro block sizes, since their element cardinality is not known in advance.
   
     These changes target management overhead rather than Avro decoding or decompression itself, reducing repeated validation, object lifecycle operations, memory reallocations, and data copying.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org