Details
-
Improvement
-
Status: Open
-
Major
-
Resolution: Unresolved
-
None
-
None
-
None
Description
This is an umbrella issue for a variety of follow-up Jiras to refactor and improve the array kernels / function machinery to have less overhead and work more efficiently for parallel processing as well as small inputs (down to ~1000 elements per kernel invocation)
Attachments
Issue Links
- is a parent of
-
ARROW-16758 [C++] Rewrite ExecuteScalarExpression to not use ScalarExecutor
- Open
-
ARROW-16756 [C++] Introduce initial ArraySpan, ExecSpan non-owning / shared_ptr-free data structures for kernel execution, refactor scalar kernels
- Resolved
-
ARROW-16824 [C++] Migrate VectorKernel implementations to use ExecSpan, ArraySpan
- Resolved
-
ARROW-16852 [C++] Migrate SCALAR_AGGREGATE, HASH_AGGREGATE functions to use ExecSpan
- Resolved
-
ARROW-16929 [C++] Remove ExecBatchIterator and usages thereof
- Resolved
- is related to
-
ARROW-16161 [C++] Overhead of std::shared_ptr<DataType> copies is causing thread contention
- Open
-
ARROW-16562 [C++] Avoid slicing array inputs in ExecBatchIterator that would result in one slice
- Closed
-
ARROW-16837 [C++] Investigate performance regressions observed in Unique, VisitArraySpanInline
- Open
-
ARROW-16845 [C++] ArraySpan::IsNull/IsValid implementations are incorrect for union types
- Open
-
ARROW-8898 [C++] Determine desirable maximum length for ExecBatch in pipelined and parallel execution of kernels
- Open
-
ARROW-13121 [C++][Compute] Extract preallocation logic from KernelExecutor
- Open
-
ARROW-16289 [C++] (eventually) abandon scalar columns of an ExecBatch in favor of RLE encoded arrays
- Open
-
ARROW-16827 [C++] Refactor internal array sorting code to use ArraySpan
- Open
-
ARROW-17129 [C++][Compute] Improve memory efficiency in Grouper
- Open
-
ARROW-9285 [C++] Detect unauthorized memory allocations in function kernels
- Open
-
ARROW-16757 [C++] Remove "scalar" output modality from array kernels
- Resolved
-
ARROW-16819 [C++] arrow::compute::CallFunction needs a batch length for nullary functions
- Resolved
-
ARROW-10026 [C++] Improve kernel performance on small batches
- Open