diff --git a/.github/workflows/build.yaml b/.github/workflows/build.yaml
index d6d3e3fdd33..b5d17022a3a 100644
--- a/.github/workflows/build.yaml
+++ b/.github/workflows/build.yaml
@@ -28,7 +28,7 @@ concurrency:
 jobs:
   cpp-build:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -37,7 +37,7 @@ jobs:
   python-build:
     needs: [cpp-build]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -46,7 +46,7 @@ jobs:
   upload-conda:
     needs: [cpp-build, python-build]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-upload-packages.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-upload-packages.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -57,7 +57,7 @@ jobs:
     if: github.ref_type == 'branch'
     needs: python-build
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       arch: "amd64"
       branch: ${{ inputs.branch }}
@@ -69,7 +69,7 @@ jobs:
       sha: ${{ inputs.sha }}
   wheel-build-libcudf:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       # build for every combination of arch and CUDA version, but only for the latest Python
       matrix_filter: group_by([.ARCH, (.CUDA_VER|split(".")|map(tonumber)|.[0])]) | map(max_by(.PY_VER|split(".")|map(tonumber)))
@@ -81,7 +81,7 @@ jobs:
   wheel-publish-libcudf:
     needs: wheel-build-libcudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -92,7 +92,7 @@ jobs:
   wheel-build-pylibcudf:
     needs: [wheel-publish-libcudf]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -102,7 +102,7 @@ jobs:
   wheel-publish-pylibcudf:
     needs: wheel-build-pylibcudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -113,7 +113,7 @@ jobs:
   wheel-build-cudf:
     needs: wheel-publish-pylibcudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -123,7 +123,7 @@ jobs:
   wheel-publish-cudf:
     needs: wheel-build-cudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -134,7 +134,7 @@ jobs:
   wheel-build-dask-cudf:
     needs: wheel-publish-cudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
       matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
@@ -146,7 +146,7 @@ jobs:
   wheel-publish-dask-cudf:
     needs: wheel-build-dask-cudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -157,7 +157,7 @@ jobs:
   wheel-build-cudf-polars:
     needs: wheel-publish-pylibcudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
       matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
@@ -169,7 +169,7 @@ jobs:
   wheel-publish-cudf-polars:
     needs: wheel-build-cudf-polars
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.10
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
diff --git a/.github/workflows/pandas-tests.yaml b/.github/workflows/pandas-tests.yaml
index d670132cca9..10c803f7921 100644
--- a/.github/workflows/pandas-tests.yaml
+++ b/.github/workflows/pandas-tests.yaml
@@ -17,7 +17,7 @@ jobs:
   pandas-tests:
       # run the Pandas unit tests
       secrets: inherit
-      uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+      uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
       with:
         # This selects "ARCH=amd64 + the latest supported Python + CUDA".
         matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
diff --git a/.github/workflows/pr.yaml b/.github/workflows/pr.yaml
index a4a8f036174..b515dbff9f3 100644
--- a/.github/workflows/pr.yaml
+++ b/.github/workflows/pr.yaml
@@ -37,7 +37,7 @@ jobs:
       - pandas-tests
       - pandas-tests-diff
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/pr-builder.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/pr-builder.yaml@branch-24.10
     if: always()
     with:
       needs: ${{ toJSON(needs) }}
@@ -52,7 +52,7 @@ jobs:
     steps:
       - name: Get PR info
         id: get-pr-info
-        uses: rapidsai/shared-actions/get-pr-info@main
+        uses: nv-gha-runners/get-pr-info@main
       - name: Checkout code repo
         uses: actions/checkout@v4
         with:
@@ -104,39 +104,39 @@ jobs:
               - '!notebooks/**'
   checks:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/checks.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/checks.yaml@branch-24.10
     with:
       enable_check_generated_files: false
   conda-cpp-build:
     needs: checks
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@branch-24.10
     with:
       build_type: pull-request
   conda-cpp-checks:
     needs: conda-cpp-build
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@branch-24.10
     with:
       build_type: pull-request
       enable_check_symbols: true
   conda-cpp-tests:
     needs: [conda-cpp-build, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@branch-24.10
     if: needs.changed-files.outputs.test_cpp == 'true'
     with:
       build_type: pull-request
   conda-python-build:
     needs: conda-cpp-build
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@branch-24.10
     with:
       build_type: pull-request
   conda-python-cudf-tests:
     needs: [conda-python-build, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@branch-24.10
     if: needs.changed-files.outputs.test_python == 'true'
     with:
       build_type: pull-request
@@ -145,7 +145,7 @@ jobs:
     # Tests for dask_cudf, custreamz, cudf_kafka are separated for CI parallelism
     needs: [conda-python-build, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@branch-24.10
     if: needs.changed-files.outputs.test_python == 'true'
     with:
       build_type: pull-request
@@ -153,7 +153,7 @@ jobs:
   conda-java-tests:
     needs: [conda-cpp-build, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     if: needs.changed-files.outputs.test_java == 'true'
     with:
       build_type: pull-request
@@ -164,7 +164,7 @@ jobs:
   static-configure:
     needs: checks
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       build_type: pull-request
       # Use the wheel container so we can skip conda solves and since our
@@ -174,7 +174,7 @@ jobs:
   conda-notebook-tests:
     needs: [conda-python-build, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     if: needs.changed-files.outputs.test_notebooks == 'true'
     with:
       build_type: pull-request
@@ -185,7 +185,7 @@ jobs:
   docs-build:
     needs: conda-python-build
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       build_type: pull-request
       node_type: "gpu-v100-latest-1"
@@ -195,7 +195,7 @@ jobs:
   wheel-build-libcudf:
     needs: checks
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       # build for every combination of arch and CUDA version, but only for the latest Python
       matrix_filter: group_by([.ARCH, (.CUDA_VER|split(".")|map(tonumber)|.[0])]) | map(max_by(.PY_VER|split(".")|map(tonumber)))
@@ -204,21 +204,21 @@ jobs:
   wheel-build-pylibcudf:
     needs: [checks, wheel-build-libcudf]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       build_type: pull-request
       script: "ci/build_wheel_pylibcudf.sh"
   wheel-build-cudf:
     needs: wheel-build-pylibcudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       build_type: pull-request
       script: "ci/build_wheel_cudf.sh"
   wheel-tests-cudf:
     needs: [wheel-build-cudf, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     if: needs.changed-files.outputs.test_python == 'true'
     with:
       build_type: pull-request
@@ -226,7 +226,7 @@ jobs:
   wheel-build-cudf-polars:
     needs: wheel-build-pylibcudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
       matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
@@ -235,7 +235,7 @@ jobs:
   wheel-tests-cudf-polars:
     needs: [wheel-build-cudf-polars, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     if: needs.changed-files.outputs.test_python == 'true'
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
@@ -247,7 +247,7 @@ jobs:
   wheel-build-dask-cudf:
     needs: wheel-build-cudf
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.10
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
       matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
@@ -256,7 +256,7 @@ jobs:
   wheel-tests-dask-cudf:
     needs: [wheel-build-dask-cudf, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     if: needs.changed-files.outputs.test_python == 'true'
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
@@ -265,7 +265,7 @@ jobs:
       script: ci/test_wheel_dask_cudf.sh
   devcontainer:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@branch-24.10
     with:
       arch: '["amd64"]'
       cuda: '["12.5"]'
@@ -276,7 +276,7 @@ jobs:
   unit-tests-cudf-pandas:
     needs: [wheel-build-cudf, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     if: needs.changed-files.outputs.test_python == 'true'
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
@@ -287,7 +287,7 @@ jobs:
     # run the Pandas unit tests using PR branch
     needs: [wheel-build-cudf, changed-files]
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     if: needs.changed-files.outputs.test_python == 'true'
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
@@ -299,7 +299,7 @@ jobs:
   pandas-tests-diff:
     # diff the results of running the Pandas unit tests and publish a job summary
     needs: pandas-tests
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
         node_type: cpu4
         build_type: pull-request
diff --git a/.github/workflows/pr_issue_status_automation.yml b/.github/workflows/pr_issue_status_automation.yml
index fe77ad4b6b2..45e5191eb54 100644
--- a/.github/workflows/pr_issue_status_automation.yml
+++ b/.github/workflows/pr_issue_status_automation.yml
@@ -23,7 +23,7 @@ on:
 
 jobs:
     get-project-id:
-      uses: rapidsai/shared-workflows/.github/workflows/project-get-item-id.yaml@python-3.12
+      uses: rapidsai/shared-workflows/.github/workflows/project-get-item-id.yaml@branch-24.10
       if: github.event.pull_request.state == 'open'
       secrets: inherit
       permissions:
@@ -34,7 +34,7 @@ jobs:
 
     update-status:
       # This job sets the PR and its linked issues to "In Progress" status
-      uses: rapidsai/shared-workflows/.github/workflows/project-get-set-single-select-field.yaml@python-3.12
+      uses: rapidsai/shared-workflows/.github/workflows/project-get-set-single-select-field.yaml@branch-24.10
       if: ${{ github.event.pull_request.state == 'open' && needs.get-project-id.outputs.ITEM_PROJECT_ID != '' }}
       needs: get-project-id
       with:
@@ -50,7 +50,7 @@ jobs:
 
     update-sprint:
       # This job sets the PR and its linked issues to the current "Weekly Sprint"
-      uses: rapidsai/shared-workflows/.github/workflows/project-get-set-iteration-field.yaml@python-3.12
+      uses: rapidsai/shared-workflows/.github/workflows/project-get-set-iteration-field.yaml@branch-24.10
       if: ${{ github.event.pull_request.state == 'open' && needs.get-project-id.outputs.ITEM_PROJECT_ID != '' }}
       needs: get-project-id
       with:
diff --git a/.github/workflows/test.yaml b/.github/workflows/test.yaml
index 4af6a0d690d..8605fa46f68 100644
--- a/.github/workflows/test.yaml
+++ b/.github/workflows/test.yaml
@@ -16,7 +16,7 @@ on:
 jobs:
   conda-cpp-checks:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -25,7 +25,7 @@ jobs:
       enable_check_symbols: true
   conda-cpp-tests:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -33,7 +33,7 @@ jobs:
       sha: ${{ inputs.sha }}
   conda-cpp-memcheck-tests:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -45,7 +45,7 @@ jobs:
       run_script: "ci/test_cpp_memcheck.sh"
   static-configure:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       build_type: pull-request
       # Use the wheel container so we can skip conda solves and since our
@@ -54,7 +54,7 @@ jobs:
       run_script: "ci/configure_cpp_static.sh"
   conda-python-cudf-tests:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -64,7 +64,7 @@ jobs:
   conda-python-other-tests:
     # Tests for dask_cudf, custreamz, cudf_kafka are separated for CI parallelism
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -73,7 +73,7 @@ jobs:
       script: "ci/test_python_other.sh"
   conda-java-tests:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -85,7 +85,7 @@ jobs:
       run_script: "ci/test_java.sh"
   conda-notebook-tests:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -97,7 +97,7 @@ jobs:
       run_script: "ci/test_notebooks.sh"
   wheel-tests-cudf:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -106,7 +106,7 @@ jobs:
       script: ci/test_wheel_cudf.sh
   wheel-tests-dask-cudf:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
       matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
@@ -117,7 +117,7 @@ jobs:
       script: ci/test_wheel_dask_cudf.sh
   unit-tests-cudf-pandas:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -126,7 +126,7 @@ jobs:
       script: ci/cudf_pandas_scripts/run_tests.sh
   third-party-integration-tests-cudf-pandas:
     secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@python-3.12
+    uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.10
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
diff --git a/cpp/benchmarks/io/parquet/parquet_reader_input.cpp b/cpp/benchmarks/io/parquet/parquet_reader_input.cpp
index 7563c823454..ce115fd7723 100644
--- a/cpp/benchmarks/io/parquet/parquet_reader_input.cpp
+++ b/cpp/benchmarks/io/parquet/parquet_reader_input.cpp
@@ -32,7 +32,8 @@ constexpr cudf::size_type num_cols = 64;
 void parquet_read_common(cudf::size_type num_rows_to_read,
                          cudf::size_type num_cols_to_read,
                          cuio_source_sink_pair& source_sink,
-                         nvbench::state& state)
+                         nvbench::state& state,
+                         size_t table_data_size = data_size)
 {
   cudf::io::parquet_reader_options read_opts =
     cudf::io::parquet_reader_options::builder(source_sink.make_source_info());
@@ -52,7 +53,7 @@ void parquet_read_common(cudf::size_type num_rows_to_read,
     });
 
   auto const time = state.get_summary("nv/cold/time/gpu/mean").get_float64("value");
-  state.add_element_count(static_cast<double>(data_size) / time, "bytes_per_second");
+  state.add_element_count(static_cast<double>(table_data_size) / time, "bytes_per_second");
   state.add_buffer_size(
     mem_stats_logger.peak_memory_usage(), "peak_memory_usage", "peak_memory_usage");
   state.add_buffer_size(source_sink.size(), "encoded_file_size", "encoded_file_size");
@@ -231,6 +232,70 @@ void BM_parquet_read_chunks(nvbench::state& state, nvbench::type_list<nvbench::e
   state.add_buffer_size(source_sink.size(), "encoded_file_size", "encoded_file_size");
 }
 
+template <data_type DataType>
+void BM_parquet_read_wide_tables(nvbench::state& state,
+                                 nvbench::type_list<nvbench::enum_type<DataType>> type_list)
+{
+  auto const d_type = get_type_or_group(static_cast<int32_t>(DataType));
+
+  auto const n_col           = static_cast<cudf::size_type>(state.get_int64("num_cols"));
+  auto const data_size_bytes = static_cast<size_t>(state.get_int64("data_size_mb") << 20);
+  auto const cardinality     = static_cast<cudf::size_type>(state.get_int64("cardinality"));
+  auto const run_length      = static_cast<cudf::size_type>(state.get_int64("run_length"));
+  auto const source_type     = io_type::DEVICE_BUFFER;
+  cuio_source_sink_pair source_sink(source_type);
+
+  auto const num_rows_written = [&]() {
+    auto const tbl = create_random_table(
+      cycle_dtypes(d_type, n_col),
+      table_size_bytes{data_size_bytes},
+      data_profile_builder().cardinality(cardinality).avg_run_length(run_length));
+    auto const view = tbl->view();
+
+    cudf::io::parquet_writer_options write_opts =
+      cudf::io::parquet_writer_options::builder(source_sink.make_sink_info(), view)
+        .compression(cudf::io::compression_type::NONE);
+    cudf::io::write_parquet(write_opts);
+    return view.num_rows();
+  }();
+
+  parquet_read_common(num_rows_written, n_col, source_sink, state, data_size_bytes);
+}
+
+void BM_parquet_read_wide_tables_mixed(nvbench::state& state)
+{
+  auto const d_type = []() {
+    auto d_type1 = get_type_or_group(static_cast<int32_t>(data_type::INTEGRAL));
+    auto d_type2 = get_type_or_group(static_cast<int32_t>(data_type::FLOAT));
+    d_type1.reserve(d_type1.size() + d_type2.size());
+    std::move(d_type2.begin(), d_type2.end(), std::back_inserter(d_type1));
+    return d_type1;
+  }();
+
+  auto const n_col           = static_cast<cudf::size_type>(state.get_int64("num_cols"));
+  auto const data_size_bytes = static_cast<size_t>(state.get_int64("data_size_mb") << 20);
+  auto const cardinality     = static_cast<cudf::size_type>(state.get_int64("cardinality"));
+  auto const run_length      = static_cast<cudf::size_type>(state.get_int64("run_length"));
+  auto const source_type     = io_type::DEVICE_BUFFER;
+  cuio_source_sink_pair source_sink(source_type);
+
+  auto const num_rows_written = [&]() {
+    auto const tbl = create_random_table(
+      cycle_dtypes(d_type, n_col),
+      table_size_bytes{data_size_bytes},
+      data_profile_builder().cardinality(cardinality).avg_run_length(run_length));
+    auto const view = tbl->view();
+
+    cudf::io::parquet_writer_options write_opts =
+      cudf::io::parquet_writer_options::builder(source_sink.make_sink_info(), view)
+        .compression(cudf::io::compression_type::NONE);
+    cudf::io::write_parquet(write_opts);
+    return view.num_rows();
+  }();
+
+  parquet_read_common(num_rows_written, n_col, source_sink, state, data_size_bytes);
+}
+
 using d_type_list = nvbench::enum_type_list<data_type::INTEGRAL,
                                             data_type::FLOAT,
                                             data_type::DECIMAL,
@@ -272,6 +337,24 @@ NVBENCH_BENCH(BM_parquet_read_io_small_mixed)
   .add_int64_axis("run_length", {1, 32})
   .add_int64_axis("num_string_cols", {1, 2, 3});
 
+using d_type_list_wide_table = nvbench::enum_type_list<data_type::DECIMAL, data_type::STRING>;
+NVBENCH_BENCH_TYPES(BM_parquet_read_wide_tables, NVBENCH_TYPE_AXES(d_type_list_wide_table))
+  .set_name("parquet_read_wide_tables")
+  .set_min_samples(4)
+  .set_type_axes_names({"data_type"})
+  .add_int64_axis("data_size_mb", {1024, 2048, 4096})
+  .add_int64_axis("num_cols", {256, 512, 1024})
+  .add_int64_axis("cardinality", {0, 1000})
+  .add_int64_axis("run_length", {1, 32});
+
+NVBENCH_BENCH(BM_parquet_read_wide_tables_mixed)
+  .set_name("parquet_read_wide_tables_mixed")
+  .set_min_samples(4)
+  .add_int64_axis("data_size_mb", {1024, 2048, 4096})
+  .add_int64_axis("num_cols", {256, 512, 1024})
+  .add_int64_axis("cardinality", {0, 1000})
+  .add_int64_axis("run_length", {1, 32});
+
 // a benchmark for structs that only contain fixed-width types
 using d_type_list_struct_only = nvbench::enum_type_list<data_type::STRUCT>;
 NVBENCH_BENCH_TYPES(BM_parquet_read_fixed_width_struct, NVBENCH_TYPE_AXES(d_type_list_struct_only))
diff --git a/cpp/include/cudf/io/parquet.hpp b/cpp/include/cudf/io/parquet.hpp
index ed7b2ac0850..ee03a382bec 100644
--- a/cpp/include/cudf/io/parquet.hpp
+++ b/cpp/include/cudf/io/parquet.hpp
@@ -39,8 +39,9 @@ namespace io {
  * @file
  */
 
-constexpr size_t default_row_group_size_bytes   = 128 * 1024 * 1024;  ///< 128MB per row group
-constexpr size_type default_row_group_size_rows = 1000000;     ///< 1 million rows per row group
+constexpr size_t default_row_group_size_bytes =
+  std::numeric_limits<size_t>::max();                          ///< Infinite bytes per row group
+constexpr size_type default_row_group_size_rows = 1'000'000;   ///< 1 million rows per row group
 constexpr size_t default_max_page_size_bytes    = 512 * 1024;  ///< 512KB per page
 constexpr size_type default_max_page_size_rows  = 20000;       ///< 20k rows per page
 constexpr int32_t default_column_index_truncate_length = 64;   ///< truncate to 64 bytes
diff --git a/cpp/src/io/parquet/writer_impl.cu b/cpp/src/io/parquet/writer_impl.cu
index 81fd4ab9f82..ec05f35d405 100644
--- a/cpp/src/io/parquet/writer_impl.cu
+++ b/cpp/src/io/parquet/writer_impl.cu
@@ -1819,8 +1819,14 @@ auto convert_table_to_parquet_data(table_input_metadata& table_meta,
     auto const table_size  = std::reduce(column_sizes.begin(), column_sizes.end());
     auto const avg_row_len = util::div_rounding_up_safe<size_t>(table_size, input.num_rows());
     if (avg_row_len > 0) {
-      auto const rg_frag_size = util::div_rounding_up_safe(max_row_group_size, avg_row_len);
-      max_page_fragment_size  = std::min<size_type>(rg_frag_size, max_page_fragment_size);
+      // Ensure `rg_frag_size` is not bigger than size_type::max for default max_row_group_size
+      // value (=uint64::max) to avoid a sign overflow when comparing
+      auto const rg_frag_size =
+        std::min<size_t>(std::numeric_limits<size_type>::max(),
+                         util::div_rounding_up_safe(max_row_group_size, avg_row_len));
+      // Safe comparison as rg_frag_size fits in size_type
+      max_page_fragment_size =
+        std::min<size_type>(static_cast<size_type>(rg_frag_size), max_page_fragment_size);
     }
 
     // dividing page size by average row length will tend to overshoot the desired
diff --git a/cpp/src/join/join_common_utils.hpp b/cpp/src/join/join_common_utils.hpp
index 86402a0e7de..573101cefd9 100644
--- a/cpp/src/join/join_common_utils.hpp
+++ b/cpp/src/join/join_common_utils.hpp
@@ -22,7 +22,6 @@
 #include <cudf/table/row_operators.cuh>
 #include <cudf/table/table_view.hpp>
 
-#include <cuco/static_map.cuh>
 #include <cuco/static_multimap.cuh>
 #include <cuda/atomic>
 
@@ -51,11 +50,6 @@ using mixed_multimap_type =
                         cudf::detail::cuco_allocator<char>,
                         cuco::legacy::double_hashing<1, hash_type, hash_type>>;
 
-using semi_map_type = cuco::legacy::static_map<hash_value_type,
-                                               size_type,
-                                               cuda::thread_scope_device,
-                                               cudf::detail::cuco_allocator<char>>;
-
 using row_hash_legacy =
   cudf::row_hasher<cudf::hashing::detail::default_hash, cudf::nullate::DYNAMIC>;
 
diff --git a/cpp/src/join/mixed_join_common_utils.cuh b/cpp/src/join/mixed_join_common_utils.cuh
index 19701816867..89c13285cfe 100644
--- a/cpp/src/join/mixed_join_common_utils.cuh
+++ b/cpp/src/join/mixed_join_common_utils.cuh
@@ -25,6 +25,7 @@
 #include <rmm/device_uvector.hpp>
 
 #include <cub/cub.cuh>
+#include <cuco/static_set.cuh>
 
 namespace cudf {
 namespace detail {
@@ -160,6 +161,38 @@ struct pair_expression_equality : public expression_equality<has_nulls> {
   }
 };
 
+/**
+ * @brief Equality comparator that composes two row_equality comparators.
+ */
+struct double_row_equality_comparator {
+  row_equality const equality_comparator;
+  row_equality const conditional_comparator;
+
+  __device__ bool operator()(size_type lhs_row_index, size_type rhs_row_index) const noexcept
+  {
+    using experimental::row::lhs_index_type;
+    using experimental::row::rhs_index_type;
+
+    return equality_comparator(lhs_index_type{lhs_row_index}, rhs_index_type{rhs_row_index}) &&
+           conditional_comparator(lhs_index_type{lhs_row_index}, rhs_index_type{rhs_row_index});
+  }
+};
+
+// A CUDA Cooperative Group of 4 threads for the hash set.
+auto constexpr DEFAULT_MIXED_JOIN_CG_SIZE = 4;
+
+// The hash set type used by mixed_semi_join with the build_table.
+using hash_set_type = cuco::static_set<size_type,
+                                       cuco::extent<size_t>,
+                                       cuda::thread_scope_device,
+                                       double_row_equality_comparator,
+                                       cuco::linear_probing<DEFAULT_MIXED_JOIN_CG_SIZE, row_hash>,
+                                       cudf::detail::cuco_allocator<char>,
+                                       cuco::storage<1>>;
+
+// The hash_set_ref_type used by mixed_semi_join kerenels for probing.
+using hash_set_ref_type = hash_set_type::ref_type<cuco::contains_tag>;
+
 }  // namespace detail
 
 }  // namespace cudf
diff --git a/cpp/src/join/mixed_join_kernels_semi.cu b/cpp/src/join/mixed_join_kernels_semi.cu
index 7459ac3e99c..f2c5ff13638 100644
--- a/cpp/src/join/mixed_join_kernels_semi.cu
+++ b/cpp/src/join/mixed_join_kernels_semi.cu
@@ -38,12 +38,16 @@ CUDF_KERNEL void __launch_bounds__(block_size)
                   table_device_view right_table,
                   table_device_view probe,
                   table_device_view build,
-                  row_hash const hash_probe,
                   row_equality const equality_probe,
-                  cudf::detail::semi_map_type::device_view hash_table_view,
+                  hash_set_ref_type set_ref,
                   cudf::device_span<bool> left_table_keep_mask,
                   cudf::ast::detail::expression_device_view device_expression_data)
 {
+  auto constexpr cg_size = hash_set_ref_type::cg_size;
+
+  auto const tile =
+    cooperative_groups::tiled_partition<cg_size>(cooperative_groups::this_thread_block());
+
   // Normally the casting of a shared memory array is used to create multiple
   // arrays of different types from the shared memory buffer, but here it is
   // used to circumvent conflicts between arrays of different types between
@@ -52,24 +56,24 @@ CUDF_KERNEL void __launch_bounds__(block_size)
   cudf::ast::detail::IntermediateDataType<has_nulls>* intermediate_storage =
     reinterpret_cast<cudf::ast::detail::IntermediateDataType<has_nulls>*>(raw_intermediate_storage);
   auto thread_intermediate_storage =
-    &intermediate_storage[threadIdx.x * device_expression_data.num_intermediates];
-
-  cudf::size_type const left_num_rows  = left_table.num_rows();
-  cudf::size_type const right_num_rows = right_table.num_rows();
-  auto const outer_num_rows            = left_num_rows;
+    &intermediate_storage[tile.meta_group_rank() * device_expression_data.num_intermediates];
 
-  cudf::size_type outer_row_index = threadIdx.x + blockIdx.x * block_size;
+  cudf::size_type const outer_num_rows = left_table.num_rows();
+  auto const outer_row_index = cudf::detail::grid_1d::global_thread_id<block_size>() / cg_size;
 
   auto evaluator = cudf::ast::detail::expression_evaluator<has_nulls>(
     left_table, right_table, device_expression_data);
 
   if (outer_row_index < outer_num_rows) {
+    // Make sure to swap_tables here as hash_set will use probe table as the left one.
+    auto constexpr swap_tables = true;
     // Figure out the number of elements for this key.
     auto equality = single_expression_equality<has_nulls>{
-      evaluator, thread_intermediate_storage, false, equality_probe};
+      evaluator, thread_intermediate_storage, swap_tables, equality_probe};
 
-    left_table_keep_mask[outer_row_index] =
-      hash_table_view.contains(outer_row_index, hash_probe, equality);
+    auto const set_ref_equality = set_ref.with_key_eq(equality);
+    auto const result           = set_ref_equality.contains(tile, outer_row_index);
+    if (tile.thread_rank() == 0) left_table_keep_mask[outer_row_index] = result;
   }
 }
 
@@ -78,9 +82,8 @@ void launch_mixed_join_semi(bool has_nulls,
                             table_device_view right_table,
                             table_device_view probe,
                             table_device_view build,
-                            row_hash const hash_probe,
                             row_equality const equality_probe,
-                            cudf::detail::semi_map_type::device_view hash_table_view,
+                            hash_set_ref_type set_ref,
                             cudf::device_span<bool> left_table_keep_mask,
                             cudf::ast::detail::expression_device_view device_expression_data,
                             detail::grid_1d const config,
@@ -94,9 +97,8 @@ void launch_mixed_join_semi(bool has_nulls,
         right_table,
         probe,
         build,
-        hash_probe,
         equality_probe,
-        hash_table_view,
+        set_ref,
         left_table_keep_mask,
         device_expression_data);
   } else {
@@ -106,9 +108,8 @@ void launch_mixed_join_semi(bool has_nulls,
         right_table,
         probe,
         build,
-        hash_probe,
         equality_probe,
-        hash_table_view,
+        set_ref,
         left_table_keep_mask,
         device_expression_data);
   }
diff --git a/cpp/src/join/mixed_join_kernels_semi.cuh b/cpp/src/join/mixed_join_kernels_semi.cuh
index 43714ffb36a..b08298e64e4 100644
--- a/cpp/src/join/mixed_join_kernels_semi.cuh
+++ b/cpp/src/join/mixed_join_kernels_semi.cuh
@@ -45,9 +45,8 @@ namespace detail {
  * @param[in] right_table The right table
  * @param[in] probe The table with which to probe the hash table for matches.
  * @param[in] build The table with which the hash table was built.
- * @param[in] hash_probe The hasher used for the probe table.
  * @param[in] equality_probe The equality comparator used when probing the hash table.
- * @param[in] hash_table_view The hash table built from `build`.
+ * @param[in] set_ref The hash table device view built from `build`.
  * @param[out] left_table_keep_mask The result of the join operation with "true" element indicating
  * the corresponding index from left table is present in output
  * @param[in] device_expression_data Container of device data required to evaluate the desired
@@ -58,9 +57,8 @@ void launch_mixed_join_semi(bool has_nulls,
                             table_device_view right_table,
                             table_device_view probe,
                             table_device_view build,
-                            row_hash const hash_probe,
                             row_equality const equality_probe,
-                            cudf::detail::semi_map_type::device_view hash_table_view,
+                            hash_set_ref_type set_ref,
                             cudf::device_span<bool> left_table_keep_mask,
                             cudf::ast::detail::expression_device_view device_expression_data,
                             detail::grid_1d const config,
diff --git a/cpp/src/join/mixed_join_semi.cu b/cpp/src/join/mixed_join_semi.cu
index cfb785e242c..719b1d47105 100644
--- a/cpp/src/join/mixed_join_semi.cu
+++ b/cpp/src/join/mixed_join_semi.cu
@@ -46,45 +46,6 @@
 namespace cudf {
 namespace detail {
 
-namespace {
-/**
- * @brief Device functor to create a pair of hash value and index for a given row.
- */
-struct make_pair_function_semi {
-  __device__ __forceinline__ cudf::detail::pair_type operator()(size_type i) const noexcept
-  {
-    // The value is irrelevant since we only ever use the hash map to check for
-    // membership of a particular row index.
-    return cuco::make_pair(static_cast<hash_value_type>(i), 0);
-  }
-};
-
-/**
- * @brief Equality comparator that composes two row_equality comparators.
- */
-class double_row_equality {
- public:
-  double_row_equality(row_equality equality_comparator, row_equality conditional_comparator)
-    : _equality_comparator{equality_comparator}, _conditional_comparator{conditional_comparator}
-  {
-  }
-
-  __device__ bool operator()(size_type lhs_row_index, size_type rhs_row_index) const noexcept
-  {
-    using experimental::row::lhs_index_type;
-    using experimental::row::rhs_index_type;
-
-    return _equality_comparator(lhs_index_type{lhs_row_index}, rhs_index_type{rhs_row_index}) &&
-           _conditional_comparator(lhs_index_type{lhs_row_index}, rhs_index_type{rhs_row_index});
-  }
-
- private:
-  row_equality _equality_comparator;
-  row_equality _conditional_comparator;
-};
-
-}  // namespace
-
 std::unique_ptr<rmm::device_uvector<size_type>> mixed_join_semi(
   table_view const& left_equality,
   table_view const& right_equality,
@@ -96,7 +57,7 @@ std::unique_ptr<rmm::device_uvector<size_type>> mixed_join_semi(
   rmm::cuda_stream_view stream,
   rmm::device_async_resource_ref mr)
 {
-  CUDF_EXPECTS((join_type != join_kind::INNER_JOIN) && (join_type != join_kind::LEFT_JOIN) &&
+  CUDF_EXPECTS((join_type != join_kind::INNER_JOIN) and (join_type != join_kind::LEFT_JOIN) and
                  (join_type != join_kind::FULL_JOIN),
                "Inner, left, and full joins should use mixed_join.");
 
@@ -137,7 +98,7 @@ std::unique_ptr<rmm::device_uvector<size_type>> mixed_join_semi(
   // output column and follow the null-supporting expression evaluation code
   // path.
   auto const has_nulls = cudf::nullate::DYNAMIC{
-    cudf::has_nulls(left_equality) || cudf::has_nulls(right_equality) ||
+    cudf::has_nulls(left_equality) or cudf::has_nulls(right_equality) or
     binary_predicate.may_evaluate_null(left_conditional, right_conditional, stream)};
 
   auto const parser = ast::detail::expression_parser{
@@ -156,27 +117,20 @@ std::unique_ptr<rmm::device_uvector<size_type>> mixed_join_semi(
   auto right_conditional_view = table_device_view::create(right_conditional, stream);
 
   auto const preprocessed_build =
-    experimental::row::equality::preprocessed_table::create(build, stream);
+    cudf::experimental::row::equality::preprocessed_table::create(build, stream);
   auto const preprocessed_probe =
-    experimental::row::equality::preprocessed_table::create(probe, stream);
+    cudf::experimental::row::equality::preprocessed_table::create(probe, stream);
   auto const row_comparator =
-    cudf::experimental::row::equality::two_table_comparator{preprocessed_probe, preprocessed_build};
+    cudf::experimental::row::equality::two_table_comparator{preprocessed_build, preprocessed_probe};
   auto const equality_probe = row_comparator.equal_to<false>(has_nulls, compare_nulls);
 
-  semi_map_type hash_table{
-    compute_hash_table_size(build.num_rows()),
-    cuco::empty_key{std::numeric_limits<hash_value_type>::max()},
-    cuco::empty_value{cudf::detail::JoinNoneValue},
-    cudf::detail::cuco_allocator<char>{rmm::mr::polymorphic_allocator<char>{}, stream},
-    stream.value()};
-
   // Create hash table containing all keys found in right table
   // TODO: To add support for nested columns we will need to flatten in many
   // places. However, this probably isn't worth adding any time soon since we
   // won't be able to support AST conditions for those types anyway.
   auto const build_nulls    = cudf::nullate::DYNAMIC{cudf::has_nulls(build)};
   auto const row_hash_build = cudf::experimental::row::hash::row_hasher{preprocessed_build};
-  auto const hash_build     = row_hash_build.device_hasher(build_nulls);
+
   // Since we may see multiple rows that are identical in the equality tables
   // but differ in the conditional tables, the equality comparator used for
   // insertion must account for both sets of tables. An alternative solution
@@ -191,20 +145,28 @@ std::unique_ptr<rmm::device_uvector<size_type>> mixed_join_semi(
   auto const equality_build_equality =
     row_comparator_build.equal_to<false>(build_nulls, compare_nulls);
   auto const preprocessed_build_condtional =
-    experimental::row::equality::preprocessed_table::create(right_conditional, stream);
+    cudf::experimental::row::equality::preprocessed_table::create(right_conditional, stream);
   auto const row_comparator_conditional_build =
     cudf::experimental::row::equality::two_table_comparator{preprocessed_build_condtional,
                                                             preprocessed_build_condtional};
   auto const equality_build_conditional =
     row_comparator_conditional_build.equal_to<false>(build_nulls, compare_nulls);
-  double_row_equality equality_build{equality_build_equality, equality_build_conditional};
-  make_pair_function_semi pair_func_build{};
 
-  auto iter = cudf::detail::make_counting_transform_iterator(0, pair_func_build);
+  hash_set_type row_set{
+    {compute_hash_table_size(build.num_rows())},
+    cuco::empty_key{JoinNoneValue},
+    {equality_build_equality, equality_build_conditional},
+    {row_hash_build.device_hasher(build_nulls)},
+    {},
+    {},
+    cudf::detail::cuco_allocator<char>{rmm::mr::polymorphic_allocator<char>{}, stream},
+    {stream.value()}};
+
+  auto iter = thrust::make_counting_iterator(0);
 
   // skip rows that are null here.
   if ((compare_nulls == null_equality::EQUAL) or (not nullable(build))) {
-    hash_table.insert(iter, iter + right_num_rows, hash_build, equality_build, stream.value());
+    row_set.insert(iter, iter + right_num_rows, stream.value());
   } else {
     thrust::counting_iterator<cudf::size_type> stencil(0);
     auto const [row_bitmask, _] =
@@ -212,18 +174,19 @@ std::unique_ptr<rmm::device_uvector<size_type>> mixed_join_semi(
     row_is_valid pred{static_cast<bitmask_type const*>(row_bitmask.data())};
 
     // insert valid rows
-    hash_table.insert_if(
-      iter, iter + right_num_rows, stencil, pred, hash_build, equality_build, stream.value());
+    row_set.insert_if(iter, iter + right_num_rows, stencil, pred, stream.value());
   }
 
-  auto hash_table_view = hash_table.get_device_view();
-
   detail::grid_1d const config(outer_num_rows, DEFAULT_JOIN_BLOCK_SIZE);
-  auto const shmem_size_per_block = parser.shmem_per_thread * config.num_threads_per_block;
+  auto const shmem_size_per_block =
+    parser.shmem_per_thread *
+    cuco::detail::int_div_ceil(config.num_threads_per_block, hash_set_type::cg_size);
 
   auto const row_hash   = cudf::experimental::row::hash::row_hasher{preprocessed_probe};
   auto const hash_probe = row_hash.device_hasher(has_nulls);
 
+  hash_set_ref_type const row_set_ref = row_set.ref(cuco::contains).with_hash_function(hash_probe);
+
   // Vector used to indicate indices from left/probe table which are present in output
   auto left_table_keep_mask = rmm::device_uvector<bool>(probe.num_rows(), stream);
 
@@ -232,9 +195,8 @@ std::unique_ptr<rmm::device_uvector<size_type>> mixed_join_semi(
                          *right_conditional_view,
                          *probe_view,
                          *build_view,
-                         hash_probe,
                          equality_probe,
-                         hash_table_view,
+                         row_set_ref,
                          cudf::device_span<bool>(left_table_keep_mask),
                          parser.device_expression_data,
                          config,
diff --git a/cpp/tests/join/mixed_join_tests.cu b/cpp/tests/join/mixed_join_tests.cu
index 6c147c8a128..08a0136700d 100644
--- a/cpp/tests/join/mixed_join_tests.cu
+++ b/cpp/tests/join/mixed_join_tests.cu
@@ -778,6 +778,21 @@ TYPED_TEST(MixedLeftSemiJoinTest, BasicEquality)
              {1});
 }
 
+TYPED_TEST(MixedLeftSemiJoinTest, MixedLeftSemiJoinGatherMap)
+{
+  auto const col_ref_left_1  = cudf::ast::column_reference(0, cudf::ast::table_reference::LEFT);
+  auto const col_ref_right_1 = cudf::ast::column_reference(0, cudf::ast::table_reference::RIGHT);
+  auto left_one_greater_right_one =
+    cudf::ast::operation(cudf::ast::ast_operator::GREATER, col_ref_left_1, col_ref_right_1);
+
+  this->test({{2, 3, 9, 0, 1, 7, 4, 6, 5, 8}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 0}},
+             {{6, 5, 9, 8, 10, 32}, {0, 1, 2, 3, 4, 5}, {7, 8, 9, 0, 1, 2}},
+             {0},
+             {1},
+             left_one_greater_right_one,
+             {2, 7, 8});
+}
+
 TYPED_TEST(MixedLeftSemiJoinTest, BasicEqualityDuplicates)
 {
   this->test({{0, 1, 2, 1}, {3, 4, 5, 6}, {10, 20, 30, 40}},
@@ -900,3 +915,18 @@ TYPED_TEST(MixedLeftAntiJoinTest, AsymmetricLeftLargerEquality)
              left_zero_eq_right_zero,
              {0, 1, 3});
 }
+
+TYPED_TEST(MixedLeftAntiJoinTest, MixedLeftAntiJoinGatherMap)
+{
+  auto const col_ref_left_1  = cudf::ast::column_reference(0, cudf::ast::table_reference::LEFT);
+  auto const col_ref_right_1 = cudf::ast::column_reference(0, cudf::ast::table_reference::RIGHT);
+  auto left_one_greater_right_one =
+    cudf::ast::operation(cudf::ast::ast_operator::GREATER, col_ref_left_1, col_ref_right_1);
+
+  this->test({{2, 3, 9, 0, 1, 7, 4, 6, 5, 8}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 0}},
+             {{6, 5, 9, 8, 10, 32}, {0, 1, 2, 3, 4, 5}, {7, 8, 9, 0, 1, 2}},
+             {0},
+             {1},
+             left_one_greater_right_one,
+             {0, 1, 3, 4, 5, 6, 9});
+}
diff --git a/docs/cudf/source/cudf_pandas/usage.md b/docs/cudf/source/cudf_pandas/usage.md
index 0398a8d7086..41838e01dd9 100644
--- a/docs/cudf/source/cudf_pandas/usage.md
+++ b/docs/cudf/source/cudf_pandas/usage.md
@@ -120,3 +120,23 @@ To profile a script being run from the command line, pass the
 ```bash
 python -m cudf.pandas --profile script.py
 ```
+
+### cudf.pandas CLI Features
+
+Several of the ways to provide input to the `python` interpreter also work with `python -m cudf.pandas`, such as the REPL, the `-c` flag, and reading from stdin.
+
+Executing `python -m cudf.pandas` with no script name will enter a REPL (read-eval-print loop) similar to the behavior of the normal `python` interpreter.
+
+The `-c` flag accepts a code string to run, like this:
+
+```bash
+$ python -m cudf.pandas -c "import pandas; print(pandas)"
+<module 'pandas' (ModuleAccelerator(fast=cudf, slow=pandas))>
+```
+
+Users can also provide code to execute from stdin, like this:
+
+```bash
+$ echo "import pandas; print(pandas)" | python -m cudf.pandas
+<module 'pandas' (ModuleAccelerator(fast=cudf, slow=pandas))>
+```
diff --git a/python/cudf/cudf/_lib/parquet.pyx b/python/cudf/cudf/_lib/parquet.pyx
index a0155671a26..e6c9d60b05b 100644
--- a/python/cudf/cudf/_lib/parquet.pyx
+++ b/python/cudf/cudf/_lib/parquet.pyx
@@ -438,7 +438,7 @@ def write_parquet(
     object statistics="ROWGROUP",
     object metadata_file_path=None,
     object int96_timestamps=False,
-    object row_group_size_bytes=_ROW_GROUP_SIZE_BYTES_DEFAULT,
+    object row_group_size_bytes=None,
     object row_group_size_rows=None,
     object max_page_size_bytes=None,
     object max_page_size_rows=None,
@@ -616,9 +616,9 @@ cdef class ParquetWriter:
         Name of the compression to use. Use ``None`` for no compression.
     statistics : {'ROWGROUP', 'PAGE', 'COLUMN', 'NONE'}, default 'ROWGROUP'
         Level at which column statistics should be included in file.
-    row_group_size_bytes: int, default 134217728
+    row_group_size_bytes: int, default ``uint64 max``
         Maximum size of each stripe of the output.
-        By default, 134217728 (128MB) will be used.
+        By default, a virtually infinite size equal to ``uint64 max`` will be used.
     row_group_size_rows: int, default 1000000
         Maximum number of rows of each stripe of the output.
         By default, 1000000 (10^6 rows) will be used.
@@ -661,11 +661,11 @@ cdef class ParquetWriter:
 
     def __cinit__(self, object filepath_or_buffer, object index=None,
                   object compression="snappy", str statistics="ROWGROUP",
-                  int row_group_size_bytes=_ROW_GROUP_SIZE_BYTES_DEFAULT,
-                  int row_group_size_rows=1000000,
-                  int max_page_size_bytes=524288,
-                  int max_page_size_rows=20000,
-                  int max_dictionary_size=1048576,
+                  size_t row_group_size_bytes=_ROW_GROUP_SIZE_BYTES_DEFAULT,
+                  size_type row_group_size_rows=1000000,
+                  size_t max_page_size_bytes=524288,
+                  size_type max_page_size_rows=20000,
+                  size_t max_dictionary_size=1048576,
                   bool use_dictionary=True,
                   bool store_schema=False):
         filepaths_or_buffers = (
diff --git a/python/cudf/cudf/core/dataframe.py b/python/cudf/cudf/core/dataframe.py
index 58a16a6d504..d73ad8225ca 100644
--- a/python/cudf/cudf/core/dataframe.py
+++ b/python/cudf/cudf/core/dataframe.py
@@ -6840,7 +6840,7 @@ def to_parquet(
         statistics="ROWGROUP",
         metadata_file_path=None,
         int96_timestamps=False,
-        row_group_size_bytes=ioutils._ROW_GROUP_SIZE_BYTES_DEFAULT,
+        row_group_size_bytes=None,
         row_group_size_rows=None,
         max_page_size_bytes=None,
         max_page_size_rows=None,
diff --git a/python/cudf/cudf/io/parquet.py b/python/cudf/cudf/io/parquet.py
index 62be7378e9e..ce99f98b559 100644
--- a/python/cudf/cudf/io/parquet.py
+++ b/python/cudf/cudf/io/parquet.py
@@ -64,7 +64,7 @@ def _write_parquet(
     statistics="ROWGROUP",
     metadata_file_path=None,
     int96_timestamps=False,
-    row_group_size_bytes=ioutils._ROW_GROUP_SIZE_BYTES_DEFAULT,
+    row_group_size_bytes=None,
     row_group_size_rows=None,
     max_page_size_bytes=None,
     max_page_size_rows=None,
@@ -149,7 +149,7 @@ def write_to_dataset(
     return_metadata=False,
     statistics="ROWGROUP",
     int96_timestamps=False,
-    row_group_size_bytes=ioutils._ROW_GROUP_SIZE_BYTES_DEFAULT,
+    row_group_size_bytes=None,
     row_group_size_rows=None,
     max_page_size_bytes=None,
     max_page_size_rows=None,
@@ -205,7 +205,7 @@ def write_to_dataset(
         If ``False``, timestamps will not be altered.
     row_group_size_bytes: integer or None, default None
         Maximum size of each stripe of the output.
-        If None, 134217728 (128MB) will be used.
+        If None, no limit on row group stripe size will be used.
     row_group_size_rows: integer or None, default None
         Maximum number of rows of each stripe of the output.
         If None, 1000000 will be used.
@@ -980,7 +980,7 @@ def to_parquet(
     statistics="ROWGROUP",
     metadata_file_path=None,
     int96_timestamps=False,
-    row_group_size_bytes=ioutils._ROW_GROUP_SIZE_BYTES_DEFAULT,
+    row_group_size_bytes=None,
     row_group_size_rows=None,
     max_page_size_bytes=None,
     max_page_size_rows=None,
diff --git a/python/cudf/cudf/pandas/__main__.py b/python/cudf/cudf/pandas/__main__.py
index 3a82829eb7a..e0d3d9101a9 100644
--- a/python/cudf/cudf/pandas/__main__.py
+++ b/python/cudf/cudf/pandas/__main__.py
@@ -10,6 +10,7 @@
 """
 
 import argparse
+import code
 import runpy
 import sys
 import tempfile
@@ -21,6 +22,8 @@
 
 @contextmanager
 def profile(function_profile, line_profile, fn):
+    if fn is None and (line_profile or function_profile):
+        raise RuntimeError("Enabling the profiler requires a script name.")
     if line_profile:
         with open(fn) as f:
             lines = f.readlines()
@@ -54,6 +57,11 @@ def main():
         dest="module",
         nargs=1,
     )
+    parser.add_argument(
+        "-c",
+        dest="cmd",
+        nargs=1,
+    )
     parser.add_argument(
         "--profile",
         action="store_true",
@@ -72,9 +80,18 @@ def main():
 
     args = parser.parse_args()
 
+    if args.cmd:
+        f = tempfile.NamedTemporaryFile(mode="w+b", suffix=".py")
+        f.write(args.cmd[0].encode())
+        f.seek(0)
+        args.args.insert(0, f.name)
+
     install()
-    with profile(args.profile, args.line_profile, args.args[0]) as fn:
-        args.args[0] = fn
+
+    script_name = args.args[0] if len(args.args) > 0 else None
+    with profile(args.profile, args.line_profile, script_name) as fn:
+        if script_name is not None:
+            args.args[0] = fn
         if args.module:
             (module,) = args.module
             # run the module passing the remaining arguments
@@ -85,6 +102,21 @@ def main():
             # Remove ourself from argv and continue
             sys.argv[:] = args.args
             runpy.run_path(args.args[0], run_name="__main__")
+        else:
+            if sys.stdin.isatty():
+                banner = f"Python {sys.version} on {sys.platform}"
+                site_import = not sys.flags.no_site
+                if site_import:
+                    cprt = 'Type "help", "copyright", "credits" or "license" for more information.'
+                    banner += "\n" + cprt
+            else:
+                # Don't show prompts or banners if stdin is not a TTY
+                sys.ps1 = ""
+                sys.ps2 = ""
+                banner = ""
+
+            # Launch an interactive interpreter
+            code.interact(banner=banner, exitmsg="")
 
 
 if __name__ == "__main__":
diff --git a/python/cudf/cudf/utils/ioutils.py b/python/cudf/cudf/utils/ioutils.py
index 1627107b57d..1180da321e6 100644
--- a/python/cudf/cudf/utils/ioutils.py
+++ b/python/cudf/cudf/utils/ioutils.py
@@ -27,7 +27,7 @@
     fsspec_parquet = None
 
 _BYTES_PER_THREAD_DEFAULT = 256 * 1024 * 1024
-_ROW_GROUP_SIZE_BYTES_DEFAULT = 128 * 1024 * 1024
+_ROW_GROUP_SIZE_BYTES_DEFAULT = np.iinfo(np.uint64).max
 
 _docstring_remote_sources = """
 - cuDF supports local and remote data stores. See configuration details for
@@ -275,10 +275,9 @@
     timestamp[us] to the int96 format, which is the number of Julian
     days and the number of nanoseconds since midnight of 1970-01-01.
     If ``False``, timestamps will not be altered.
-row_group_size_bytes: integer, default {row_group_size_bytes_val}
+row_group_size_bytes: integer, default None
     Maximum size of each stripe of the output.
-    If None, {row_group_size_bytes_val}
-    ({row_group_size_bytes_val_in_mb} MB) will be used.
+    If None, no limit on row group stripe size will be used.
 row_group_size_rows: integer or None, default None
     Maximum number of rows of each stripe of the output.
     If None, 1000000 will be used.
@@ -346,10 +345,7 @@
 See Also
 --------
 cudf.read_parquet
-""".format(
-    row_group_size_bytes_val=_ROW_GROUP_SIZE_BYTES_DEFAULT,
-    row_group_size_bytes_val_in_mb=_ROW_GROUP_SIZE_BYTES_DEFAULT / 1024 / 1024,
-)
+"""
 doc_to_parquet = docfmt_partial(docstring=_docstring_to_parquet)
 
 _docstring_merge_parquet_filemetadata = """
diff --git a/python/cudf/cudf_pandas_tests/test_main.py b/python/cudf/cudf_pandas_tests/test_main.py
new file mode 100644
index 00000000000..326224c8fc0
--- /dev/null
+++ b/python/cudf/cudf_pandas_tests/test_main.py
@@ -0,0 +1,100 @@
+# SPDX-FileCopyrightText: Copyright (c) 2024 NVIDIA CORPORATION & AFFILIATES.
+# All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+import subprocess
+import tempfile
+import textwrap
+
+
+def _run_python(*, cudf_pandas, command):
+    executable = "python "
+    if cudf_pandas:
+        executable += "-m cudf.pandas "
+    return subprocess.run(
+        executable + command,
+        shell=True,
+        capture_output=True,
+        check=True,
+        text=True,
+    )
+
+
+def test_run_cudf_pandas_with_script():
+    with tempfile.NamedTemporaryFile(mode="w", suffix=".py", delete=True) as f:
+        code = textwrap.dedent(
+            """
+            import pandas as pd
+            df = pd.DataFrame({'a': [1, 2, 3]})
+            print(df['a'].sum())
+            """
+        )
+        f.write(code)
+        f.flush()
+
+        res = _run_python(cudf_pandas=True, command=f.name)
+        expect = _run_python(cudf_pandas=False, command=f.name)
+
+    assert res.stdout != ""
+    assert res.stdout == expect.stdout
+
+
+def test_run_cudf_pandas_with_script_with_cmd_args():
+    input_args_and_code = """-c 'import pandas as pd; df = pd.DataFrame({"a": [1, 2, 3]}); print(df["a"].sum())'"""
+
+    res = _run_python(cudf_pandas=True, command=input_args_and_code)
+    expect = _run_python(cudf_pandas=False, command=input_args_and_code)
+
+    assert res.stdout != ""
+    assert res.stdout == expect.stdout
+
+
+def test_run_cudf_pandas_with_script_with_cmd_args_check_cudf():
+    """Verify that cudf is active with -m cudf.pandas."""
+    input_args_and_code = """-c 'import pandas as pd; print(pd)'"""
+
+    res = _run_python(cudf_pandas=True, command=input_args_and_code)
+    expect = _run_python(cudf_pandas=False, command=input_args_and_code)
+
+    assert "cudf" in res.stdout
+    assert "cudf" not in expect.stdout
+
+
+def test_cudf_pandas_script_repl():
+    def start_repl_process(cmd):
+        return subprocess.Popen(
+            cmd.split(),
+            stdin=subprocess.PIPE,
+            stdout=subprocess.PIPE,
+            text=True,
+        )
+
+    def get_repl_output(process, commands):
+        for command in commands:
+            process.stdin.write(command)
+            process.stdin.flush()
+        return process.communicate()
+
+    p1 = start_repl_process("python -m cudf.pandas")
+    p2 = start_repl_process("python")
+    commands = [
+        "import pandas as pd\n",
+        "print(pd.Series(range(2)).sum())\n",
+        "print(pd.Series(range(5)).sum())\n",
+        "import sys\n",
+        "print(pd.Series(list('abcd')), out=sys.stderr)\n",
+    ]
+
+    res = get_repl_output(p1, commands)
+    expect = get_repl_output(p2, commands)
+
+    # Check stdout
+    assert res[0] != ""
+    assert res[0] == expect[0]
+
+    # Check stderr
+    assert res[1] != ""
+    assert res[1] == expect[1]
+
+    p1.kill()
+    p2.kill()
diff --git a/python/dask_cudf/dask_cudf/io/parquet.py b/python/dask_cudf/dask_cudf/io/parquet.py
index e793d4381d1..a781b8242fe 100644
--- a/python/dask_cudf/dask_cudf/io/parquet.py
+++ b/python/dask_cudf/dask_cudf/io/parquet.py
@@ -23,7 +23,6 @@
 from cudf.io import write_to_dataset
 from cudf.io.parquet import _apply_post_filters, _normalize_filters
 from cudf.utils.dtypes import cudf_dtype_from_pa_type
-from cudf.utils.ioutils import _ROW_GROUP_SIZE_BYTES_DEFAULT
 
 
 class CudfEngine(ArrowDatasetEngine):
@@ -341,9 +340,7 @@ def write_partition(
                 return_metadata=return_metadata,
                 statistics=kwargs.get("statistics", "ROWGROUP"),
                 int96_timestamps=kwargs.get("int96_timestamps", False),
-                row_group_size_bytes=kwargs.get(
-                    "row_group_size_bytes", _ROW_GROUP_SIZE_BYTES_DEFAULT
-                ),
+                row_group_size_bytes=kwargs.get("row_group_size_bytes", None),
                 row_group_size_rows=kwargs.get("row_group_size_rows", None),
                 max_page_size_bytes=kwargs.get("max_page_size_bytes", None),
                 max_page_size_rows=kwargs.get("max_page_size_rows", None),
@@ -365,7 +362,7 @@ def write_partition(
                     statistics=kwargs.get("statistics", "ROWGROUP"),
                     int96_timestamps=kwargs.get("int96_timestamps", False),
                     row_group_size_bytes=kwargs.get(
-                        "row_group_size_bytes", _ROW_GROUP_SIZE_BYTES_DEFAULT
+                        "row_group_size_bytes", None
                     ),
                     row_group_size_rows=kwargs.get(
                         "row_group_size_rows", None