update branch

.devcontainer/cuda12.2-conda/devcontainer.json → .devcontainer/cuda12.5-conda/devcontainer.json

@@ -3,15 +3,15 @@
     "context": "${localWorkspaceFolder}/.devcontainer",
     "dockerfile": "${localWorkspaceFolder}/.devcontainer/Dockerfile",
     "args": {
-      "CUDA": "12.2",
+      "CUDA": "12.5",
       "PYTHON_PACKAGE_MANAGER": "conda",
       "BASE": "rapidsai/devcontainers:24.08-cpp-mambaforge-ubuntu22.04"
     }
   },
   "runArgs": [
     "--rm",
     "--name",
-    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.2-conda"
+    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.5-conda"
   ],
   "hostRequirements": {"gpu": "optional"},
   "features": {
@@ -20,7 +20,7 @@
   "overrideFeatureInstallOrder": [
     "ghcr.io/rapidsai/devcontainers/features/rapids-build-utils"
   ],
-  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config,conda/pkgs,conda/${localWorkspaceFolderBasename}-cuda12.2-envs}"],
+  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config,conda/pkgs,conda/${localWorkspaceFolderBasename}-cuda12.5-envs}"],
   "postAttachCommand": ["/bin/bash", "-c", "if [ ${CODESPACES:-false} = 'true' ]; then . devcontainer-utils-post-attach-command; . rapids-post-attach-command; fi"],
   "workspaceFolder": "/home/coder",
   "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/cugraph,type=bind,consistency=consistent",
@@ -29,7 +29,7 @@
     "source=${localWorkspaceFolder}/../.cache,target=/home/coder/.cache,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.config,target=/home/coder/.config,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.conda/pkgs,target=/home/coder/.conda/pkgs,type=bind,consistency=consistent",
-    "source=${localWorkspaceFolder}/../.conda/${localWorkspaceFolderBasename}-cuda12.2-envs,target=/home/coder/.conda/envs,type=bind,consistency=consistent"
+    "source=${localWorkspaceFolder}/../.conda/${localWorkspaceFolderBasename}-cuda12.5-envs,target=/home/coder/.conda/envs,type=bind,consistency=consistent"
   ],
   "customizations": {
     "vscode": {

.devcontainer/cuda12.2-pip/devcontainer.json → .devcontainer/cuda12.5-pip/devcontainer.json

@@ -3,20 +3,20 @@
     "context": "${localWorkspaceFolder}/.devcontainer",
     "dockerfile": "${localWorkspaceFolder}/.devcontainer/Dockerfile",
     "args": {
-      "CUDA": "12.2",
+      "CUDA": "12.5",
       "PYTHON_PACKAGE_MANAGER": "pip",
-      "BASE": "rapidsai/devcontainers:24.08-cpp-cuda12.2-ucx1.15.0-openmpi-ubuntu22.04"
+      "BASE": "rapidsai/devcontainers:24.08-cpp-cuda12.5-ucx1.15.0-openmpi-ubuntu22.04"
     }
   },
   "runArgs": [
     "--rm",
     "--name",
-    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.2-pip"
+    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.5-pip"
   ],
   "hostRequirements": {"gpu": "optional"},
   "features": {
     "ghcr.io/rapidsai/devcontainers/features/cuda:24.8": {
-      "version": "12.2",
+      "version": "12.5",
       "installcuBLAS": true,
       "installcuSOLVER": true,
       "installcuRAND": true,
@@ -28,15 +28,15 @@
     "ghcr.io/rapidsai/devcontainers/features/cuda",
     "ghcr.io/rapidsai/devcontainers/features/rapids-build-utils"
   ],
-  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config/pip,local/share/${localWorkspaceFolderBasename}-cuda12.2-venvs}"],
+  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config/pip,local/share/${localWorkspaceFolderBasename}-cuda12.5-venvs}"],
   "postAttachCommand": ["/bin/bash", "-c", "if [ ${CODESPACES:-false} = 'true' ]; then . devcontainer-utils-post-attach-command; . rapids-post-attach-command; fi"],
   "workspaceFolder": "/home/coder",
   "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/cugraph,type=bind,consistency=consistent",
   "mounts": [
     "source=${localWorkspaceFolder}/../.aws,target=/home/coder/.aws,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.cache,target=/home/coder/.cache,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.config,target=/home/coder/.config,type=bind,consistency=consistent",
-    "source=${localWorkspaceFolder}/../.local/share/${localWorkspaceFolderBasename}-cuda12.2-venvs,target=/home/coder/.local/share/venvs,type=bind,consistency=consistent"
+    "source=${localWorkspaceFolder}/../.local/share/${localWorkspaceFolderBasename}-cuda12.5-venvs,target=/home/coder/.local/share/venvs,type=bind,consistency=consistent"
   ],
   "customizations": {
     "vscode": {

.github/workflows/pr.yaml

@@ -191,7 +191,7 @@ jobs:
     uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@branch-24.08
     with:
       arch: '["amd64"]'
-      cuda: '["12.2"]'
+      cuda: '["12.5"]'
       node_type: cpu32
       extra-repo-deploy-key: CUGRAPH_OPS_SSH_PRIVATE_DEPLOY_KEY
       build_command: |

benchmarks/cugraph/pytest-based/bench_algos.py

@@ -330,8 +330,17 @@ def bench_bfs(gpubenchmark, graph):
 
 
 def bench_sssp(gpubenchmark, graph):
+    if not graph.is_weighted():
+        pytest.skip("Skipping: Unweighted Graphs are not supported by SSSP")
+
     sssp = dask_cugraph.sssp if is_graph_distributed(graph) else cugraph.sssp
-    start = graph.edgelist.edgelist_df["src"][0]
+
+    start_col = graph.select_random_vertices(num_vertices=1)
+    if is_graph_distributed(graph):
+        start_col = start_col.compute()
+
+    start = start_col.to_arrow().to_pylist()[0]
+
     gpubenchmark(sssp, graph, start)
 
 

benchmarks/pytest.ini

@@ -1,67 +1,67 @@
 [pytest]
 pythonpath =
-             shared/python
+    shared/python
 
 testpaths =
-            cugraph/pytest_based
-	    cugraph-service/pytest_based
+    cugraph/pytest_based
+    cugraph-service/pytest_based
 
 addopts =
-          --benchmark-columns="min, max, mean, stddev, outliers"
+    --benchmark-columns="min, max, mean, stddev, outliers"
 
 markers =
-          managedmem_on: RMM managed memory enabled
-          managedmem_off: RMM managed memory disabled
-          poolallocator_on: RMM pool allocator enabled
-          poolallocator_off: RMM pool allocator disabled
-          tiny: tiny datasets
-          small: small datasets
-	  medium: medium datasets
-          large: large datasets
-          directed: directed datasets
-          undirected: undirected datasets
-	  matrix_types: inputs are matrices
-	  nx_types: inputs are NetowrkX Graph objects
-	  cugraph_types: inputs are cuGraph Graph objects
-	  sg: single-GPU
-	  mg: multi-GPU
-	  snmg: single-node multi-GPU
-	  mnmg: multi-node multi-GPU
-	  local: local cugraph
-	  remote: cugraph-service
-	  batch_size_100: batch size of 100 for sampling algos
-	  batch_size_500: batch size of 500 for sampling algos
-	  batch_size_1000: batch size of 1000 for sampling algos
-	  batch_size_2500: batch size of 2500 for sampling algos
-	  batch_size_5000: batch size of 5000 for sampling algos
-	  batch_size_10000: batch size of 10000 for sampling algos
-	  batch_size_20000: batch size of 20000 for sampling algos
-	  batch_size_30000: batch size of 30000 for sampling algos
-	  batch_size_40000: batch size of 40000 for sampling algos
-	  batch_size_50000: batch size of 50000 for sampling algos
-	  batch_size_60000: batch size of 60000 for sampling algos
-	  batch_size_70000: batch size of 70000 for sampling algos
-	  batch_size_80000: batch size of 80000 for sampling algos
-	  batch_size_90000: batch size of 90000 for sampling algos
-	  batch_size_100000: batch size of 100000 for sampling algos
-	  num_clients_2: start 2 cugraph-service clients
-	  num_clients_4: start 4 cugraph-service clients
-	  num_clients_8: start 8 cugraph-service clients
-	  num_clients_16: start 16 cugraph-service clients
-	  num_clients_32: start 32 cugraph-service clients
-	  fanout_10_25: fanout [10, 25] for sampling algos
-	  fanout_5_10_15: fanout [5, 10, 15] for sampling algos
-	  rmat_data: RMAT-generated synthetic datasets
-	  file_data: datasets from $RAPIDS_DATASET_ROOT_DIR
+    managedmem_on: RMM managed memory enabled
+    managedmem_off: RMM managed memory disabled
+    poolallocator_on: RMM pool allocator enabled
+    poolallocator_off: RMM pool allocator disabled
+    tiny: tiny datasets
+    small: small datasets
+    medium: medium datasets
+    large: large datasets
+    directed: directed datasets
+    undirected: undirected datasets
+    matrix_types: inputs are matrices
+    nx_types: inputs are NetowrkX Graph objects
+    cugraph_types: inputs are cuGraph Graph objects
+    sg: single-GPU
+    mg: multi-GPU
+    snmg: single-node multi-GPU
+    mnmg: multi-node multi-GPU
+    local: local cugraph
+    remote: cugraph-service
+    batch_size_100: batch size of 100 for sampling algos
+    batch_size_500: batch size of 500 for sampling algos
+    batch_size_1000: batch size of 1000 for sampling algos
+    batch_size_2500: batch size of 2500 for sampling algos
+    batch_size_5000: batch size of 5000 for sampling algos
+    batch_size_10000: batch size of 10000 for sampling algos
+    batch_size_20000: batch size of 20000 for sampling algos
+    batch_size_30000: batch size of 30000 for sampling algos
+    batch_size_40000: batch size of 40000 for sampling algos
+    batch_size_50000: batch size of 50000 for sampling algos
+    batch_size_60000: batch size of 60000 for sampling algos
+    batch_size_70000: batch size of 70000 for sampling algos
+    batch_size_80000: batch size of 80000 for sampling algos
+    batch_size_90000: batch size of 90000 for sampling algos
+    batch_size_100000: batch size of 100000 for sampling algos
+    num_clients_2: start 2 cugraph-service clients
+    num_clients_4: start 4 cugraph-service clients
+    num_clients_8: start 8 cugraph-service clients
+    num_clients_16: start 16 cugraph-service clients
+    num_clients_32: start 32 cugraph-service clients
+    fanout_10_25: fanout [10, 25] for sampling algos
+    fanout_5_10_15: fanout [5, 10, 15] for sampling algos
+    rmat_data: RMAT-generated synthetic datasets
+    file_data: datasets from $RAPIDS_DATASET_ROOT_DIR
 
 python_classes =
-                 Bench*
-                 Test*
+    Bench*
+    Test*
 
 python_files =
-                 bench_*
-                 test_*
+    bench_*
+    test_*
 
 python_functions =
-                   bench_*
-                   test_*
+    bench_*
+    test_*

conda/environments/all_cuda-118_arch-x86_64.yaml

@@ -71,7 +71,7 @@ dependencies:
 - sphinx-markdown-tables
 - sphinx<6
 - sphinxcontrib-websupport
-- thriftpy2<=0.5.0
+- thriftpy2!=0.5.0,!=0.5.1
 - ucx-proc=*=gpu
 - ucx-py==0.39.*,>=0.0.0a0
 - wget

conda/environments/all_cuda-122_arch-x86_64.yaml → conda/environments/all_cuda-125_arch-x86_64.yaml

@@ -17,7 +17,7 @@ dependencies:
 - cuda-nvcc
 - cuda-nvtx-dev
 - cuda-profiler-api
-- cuda-version=12.2
+- cuda-version=12.5
 - cudf==24.8.*,>=0.0.0a0
 - cupy>=12.0.0
 - cxx-compiler
@@ -76,9 +76,9 @@ dependencies:
 - sphinx-markdown-tables
 - sphinx<6
 - sphinxcontrib-websupport
-- thriftpy2<=0.5.0
+- thriftpy2!=0.5.0,!=0.5.1
 - ucx-proc=*=gpu
 - ucx-py==0.39.*,>=0.0.0a0
 - wget
 - wheel
-name: all_cuda-122_arch-x86_64
+name: all_cuda-125_arch-x86_64

conda/recipes/cugraph-service/meta.yaml

@@ -32,7 +32,7 @@ outputs:
         - rapids-build-backend>=0.3.1,<0.4.0.dev0
       run:
         - python
-        - thriftpy2 >=0.4.15
+        - thriftpy2 >=0.4.15,!=0.5.0,!=0.5.1
 
   - name: cugraph-service-server
     version: {{ version }}
@@ -65,7 +65,7 @@ outputs:
         - numpy >=1.23,<2.0a0
         - python
         - rapids-dask-dependency ={{ minor_version }}
-        - thriftpy2 >=0.4.15
+        - thriftpy2 >=0.4.15,!=0.5.0,!=0.5.1
         - ucx-py {{ ucx_py_version }}
 
 tests:

cpp/CMakeLists.txt

@@ -652,7 +652,7 @@ add_library(cugraph_c
         src/c_api/lookup_src_dst.cpp
         src/c_api/louvain.cpp
         src/c_api/triangle_count.cpp
-        src/c_api/uniform_neighbor_sampling.cpp
+        src/c_api/neighbor_sampling.cpp
         src/c_api/labeling_result.cpp
         src/c_api/weakly_connected_components.cpp
         src/c_api/strongly_connected_components.cpp

cpp/include/cugraph/graph_functions.hpp

@@ -730,6 +730,72 @@ create_graph_from_edgelist(raft::handle_t const& handle,
                            bool renumber,
                            bool do_expensive_check = false);
 
+/**
+ * @brief create a graph from (the optional vertex list and) the given edge list (with optional edge
+ * IDs and types).
+ *
+ * This version takes edge list in multiple chunks (e.g. edge data from multiple files).
+ *
+ * @tparam vertex_t Type of vertex identifiers. Needs to be an integral type.
+ * @tparam edge_t Type of edge identifiers. Needs to be an integral type.
+ * @tparam weight_t Type of edge weight.  Needs to be floating point type
+ * @tparam edge_id_t Type of edge id.  Needs to be an integral type
+ * @tparam edge_type_t Type of edge type.  Needs to be an integral type, currently only int32_t is
+ * supported
+ * @tparam store_transposed Flag indicating whether to use sources (if false) or destinations (if
+ * true) as major indices in storing edges using a 2D sparse matrix. transposed.
+ * @tparam multi_gpu Flag indicating whether template instantiation should target single-GPU (false)
+ * or multi-GPU (true).
+ * @param handle RAFT handle object to encapsulate resources (e.g. CUDA stream, communicator, and
+ * handles to various CUDA libraries) to run graph algorithms.
+ * @param vertices  If valid, part of the entire set of vertices in the graph to be renumbered.
+ * This parameter can be used to include isolated vertices. If @p renumber is false and @p vertices
+ * is valid, @p vertices elements should be consecutive integers starting from 0. If multi-GPU,
+ * applying the compute_gpu_id_from_vertex_t to every vertex should return the local GPU ID for this
+ * function to work (vertices should be pre-shuffled).
+ * @param edgelist_srcs Vectors of edge source vertex IDs. If multi-GPU, applying the
+ * compute_gpu_id_from_ext_edge_endpoints_t to every edge should return the local GPU ID for this
+ * function to work (edges should be pre-shuffled).
+ * @param edgelist_dsts Vectors of edge destination vertex IDs.
+ * @param edgelist_weights Vectors of weight values for edges
+ * @param edgelist_edge_ids Vectors of edge_id values for edges
+ * @param edgelist_edge_types Vectors of edge_type values for edges
+ * @param graph_properties Properties of the graph represented by the input (optional vertex list
+ * and) edge list.
+ * @param renumber Flag indicating whether to renumber vertices or not (must be true if @p multi_gpu
+ * is true).
+ * @param do_expensive_check A flag to run expensive checks for input arguments (if set to `true`).
+ * @return Tuple of the generated graph and optional edge_property_t objects storing the provided
+ * edge properties and a renumber map (if @p renumber is true).
+ */
+template <typename vertex_t,
+          typename edge_t,
+          typename weight_t,
+          typename edge_id_t,
+          typename edge_type_t,
+          bool store_transposed,
+          bool multi_gpu>
+std::tuple<
+  graph_t<vertex_t, edge_t, store_transposed, multi_gpu>,
+  std::optional<
+    edge_property_t<graph_view_t<vertex_t, edge_t, store_transposed, multi_gpu>, weight_t>>,
+  std::optional<
+    edge_property_t<graph_view_t<vertex_t, edge_t, store_transposed, multi_gpu>, edge_id_t>>,
+  std::optional<
+    edge_property_t<graph_view_t<vertex_t, edge_t, store_transposed, multi_gpu>, edge_type_t>>,
+  std::optional<rmm::device_uvector<vertex_t>>>
+create_graph_from_edgelist(
+  raft::handle_t const& handle,
+  std::optional<rmm::device_uvector<vertex_t>>&& vertices,
+  std::vector<rmm::device_uvector<vertex_t>>&& edgelist_srcs,
+  std::vector<rmm::device_uvector<vertex_t>>&& edgelist_dsts,
+  std::optional<std::vector<rmm::device_uvector<weight_t>>>&& edgelist_weights,
+  std::optional<std::vector<rmm::device_uvector<edge_id_t>>>&& edgelist_edge_ids,
+  std::optional<std::vector<rmm::device_uvector<edge_type_t>>>&& edgelist_edge_types,
+  graph_properties_t graph_properties,
+  bool renumber,
+  bool do_expensive_check = false);
+
 /**
  * @brief      Find all 2-hop neighbors in the graph
  *