Jason Lowe
Jason Lowe
The `ArrayMin` expression has a `withPsNote` annotation, but this annotation does not appear in the generated documentation at https://github.com/NVIDIA/spark-rapids/blob/branch-22.10/docs/supported_ops.md.
#6298 adds BinaryType support for many operations, but there were some operations left out, such as aggregations, window operations, etc. Some of these should be relatively straightforward to support, such...
**Is your feature request related to a problem? Please describe.** Apache Spark internally tracks timestamps relative to the epoch in the JVM's default timezone. cudf uses timestamps relative to the...
**Describe the bug** ContiguousSplitUntypedTest fails when run with the arena allocator but passes when run with the pool or cuda allocators: ``` [ RUN ] ContiguousSplitUntypedTest.CalculationOverflow unknown file: Failure C++...
The RAPIDS Accelerator should support GPU acceleration of shuffles that partition on `ArrayType`.
It is not always clear which configs are changeable at runtime and which are only examined on startup. The config documentation ideally should explicitly call out all configs whose values...
Adds a few more settings to the Spark defaults that are updated as part of setting up the cluster to run with the RAPIDS Accelerator. We have found these settings...
**Describe the bug** Using libcudf to load a Parquet file that is malformed "succeeds" by producing a table with some corrupted rows rather than returning an error as expected. Spark...
**Describe the bug** Attempting to load a non-existent column from a Parquet file throws ArrayIndexOutOfBoundsException instead of a more specific error. **Steps/Code to reproduce bug** Try to load a single...
**Is your feature request related to a problem? Please describe.** It's currently complicated to setup and collect an Nsight Systems trace of one or more executors, especially in non-standalone environments....