Merge remote-tracking branch 'upstream/dev' into retrieve

Author: sleeepyjack

.devcontainer/cuda11.8-gcc11/devcontainer.json

@@ -1,6 +1,6 @@
 {
   "shutdownAction": "stopContainer",
-  "image": "rapidsai/devcontainers:24.06-cpp-gcc11-cuda11.8-ubuntu22.04",
+  "image": "rapidsai/devcontainers:24.10-cpp-gcc11-cuda11.8-ubuntu22.04",
   "hostRequirements": {
     "gpu": true
   },

.devcontainer/cuda12.5-gcc12/devcontainer.json

@@ -0,0 +1,42 @@
+{
+  "shutdownAction": "stopContainer",
+  "image": "rapidsai/devcontainers:24.10-cpp-gcc12-cuda12.5-ubuntu22.04",
+  "hostRequirements": {
+    "gpu": true
+  },
+  "initializeCommand": [
+    "/bin/bash",
+    "-c",
+    "mkdir -m 0755 -p ${localWorkspaceFolder}/.{aws,cache,config}"
+  ],
+  "containerEnv": {
+    "SCCACHE_REGION": "us-east-2",
+    "SCCACHE_BUCKET": "rapids-sccache-devs",
+    "VAULT_HOST": "https://vault.ops.k8s.rapids.ai",
+    "HISTFILE": "${containerWorkspaceFolder}/.cache/._bash_history",
+    "DEVCONTAINER_NAME": "cuda12.5-gcc12",
+    "CUCO_CUDA_VERSION": "12.5",
+    "CUCO_HOST_COMPILER": "gcc",
+    "CUCO_HOST_COMPILER_VERSION": "12"
+  },
+  "workspaceFolder": "/home/coder/${localWorkspaceFolderBasename}",
+  "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/${localWorkspaceFolderBasename},type=bind,consistency=consistent",
+  "mounts": [
+    "source=${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"
+  ],
+  "customizations": {
+    "vscode": {
+      "extensions": [
+        "llvm-vs-code-extensions.vscode-clangd"
+      ],
+      "settings": {
+        "clangd.arguments": [
+          "--compile-commands-dir=${workspaceFolder}/build/latest"
+        ]
+      }
+    }
+  },
+  "name": "cuda12.5-gcc12"
+}

.devcontainer/cuda12.5-gcc13/devcontainer.json

@@ -0,0 +1,42 @@
+{
+  "shutdownAction": "stopContainer",
+  "image": "rapidsai/devcontainers:24.10-cpp-gcc13-cuda12.5-ubuntu22.04",
+  "hostRequirements": {
+    "gpu": true
+  },
+  "initializeCommand": [
+    "/bin/bash",
+    "-c",
+    "mkdir -m 0755 -p ${localWorkspaceFolder}/.{aws,cache,config}"
+  ],
+  "containerEnv": {
+    "SCCACHE_REGION": "us-east-2",
+    "SCCACHE_BUCKET": "rapids-sccache-devs",
+    "VAULT_HOST": "https://vault.ops.k8s.rapids.ai",
+    "HISTFILE": "${containerWorkspaceFolder}/.cache/._bash_history",
+    "DEVCONTAINER_NAME": "cuda12.5-gcc13",
+    "CUCO_CUDA_VERSION": "12.5",
+    "CUCO_HOST_COMPILER": "gcc",
+    "CUCO_HOST_COMPILER_VERSION": "13"
+  },
+  "workspaceFolder": "/home/coder/${localWorkspaceFolderBasename}",
+  "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/${localWorkspaceFolderBasename},type=bind,consistency=consistent",
+  "mounts": [
+    "source=${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"
+  ],
+  "customizations": {
+    "vscode": {
+      "extensions": [
+        "llvm-vs-code-extensions.vscode-clangd"
+      ],
+      "settings": {
+        "clangd.arguments": [
+          "--compile-commands-dir=${workspaceFolder}/build/latest"
+        ]
+      }
+    }
+  },
+  "name": "cuda12.5-gcc13"
+}

.devcontainer/devcontainer.json

@@ -1,6 +1,6 @@
 {
   "shutdownAction": "stopContainer",
-  "image": "rapidsai/devcontainers:24.06-cpp-gcc12-cuda12.4-ubuntu22.04",
+  "image": "rapidsai/devcontainers:24.10-cpp-gcc13-cuda12.5-ubuntu22.04",
   "hostRequirements": {
     "gpu": true
   },
@@ -14,10 +14,10 @@
     "SCCACHE_BUCKET": "rapids-sccache-devs",
     "VAULT_HOST": "https://vault.ops.k8s.rapids.ai",
     "HISTFILE": "${containerWorkspaceFolder}/.cache/._bash_history",
-    "DEVCONTAINER_NAME": "cuda12.4-gcc12",
-    "CUCO_CUDA_VERSION": "12.4",
+    "DEVCONTAINER_NAME": "cuda12.5-gcc13",
+    "CUCO_CUDA_VERSION": "12.5",
     "CUCO_HOST_COMPILER": "gcc",
-    "CUCO_HOST_COMPILER_VERSION": "12"
+    "CUCO_HOST_COMPILER_VERSION": "13"
   },
   "workspaceFolder": "/home/coder/${localWorkspaceFolderBasename}",
   "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/${localWorkspaceFolderBasename},type=bind,consistency=consistent",
@@ -38,5 +38,5 @@
       }
     }
   },
-  "name": "cuda12.4-gcc12"
+  "name": "cuda12.5-gcc13"
 }

CMakeLists.txt

@@ -16,7 +16,7 @@
 cmake_minimum_required(VERSION 3.23.1 FATAL_ERROR)
 
 if(NOT EXISTS ${CMAKE_CURRENT_BINARY_DIR}/CUCO_RAPIDS.cmake)
-    file(DOWNLOAD https://raw.githubusercontent.com/rapidsai/rapids-cmake/branch-24.08/RAPIDS.cmake
+    file(DOWNLOAD https://raw.githubusercontent.com/rapidsai/rapids-cmake/branch-24.10/RAPIDS.cmake
          ${CMAKE_CURRENT_BINARY_DIR}/CUCO_RAPIDS.cmake)
 endif()
 include(${CMAKE_CURRENT_BINARY_DIR}/CUCO_RAPIDS.cmake)

README.md

@@ -206,8 +206,8 @@ We plan to add many GPU-accelerated, concurrent data structures to `cuCollection
 #### Examples:
 - [Host-bulk APIs](https://github.com/NVIDIA/cuCollections/blob/dev/examples/static_set/host_bulk_example.cu) (see [live example in godbolt](https://godbolt.org/z/96re4zhjo))
 - [Device-ref APIs for individual operations](https://github.com/NVIDIA/cuCollections/blob/dev/examples/static_set/device_ref_example.cu) (see [live example in godbolt](https://godbolt.org/z/7aKWdGTfx))
-- [One single storage for multiple sets](https://github.com/NVIDIA/cuCollections/blob/dev/examples/static_set/device_subsets_example.cu) (see [live example in godbolt](https://godbolt.org/z/sMfqGxdha))
-- [Using shared memory as storage](https://github.com/NVIDIA/cuCollections/blob/dev/examples/static_set/shared_memory_example.cu) (see [live example in godbolt](https://godbolt.org/z/zdTnbE1q5))
+- [One single storage for multiple sets](https://github.com/NVIDIA/cuCollections/blob/dev/examples/static_set/device_subsets_example.cu) (see [live example in godbolt](https://godbolt.org/z/7f9KW44P4))
+- [Using shared memory as storage](https://github.com/NVIDIA/cuCollections/blob/dev/examples/static_set/shared_memory_example.cu) (see [live example in godbolt](https://godbolt.org/z/Ws5c71T4z))
 - [Using set as mapping table to handle large keys or indeterministic sentinels](https://github.com/NVIDIA/cuCollections/blob/dev/examples/static_set/mapping_table_example.cu) (see [live example in godbolt](https://godbolt.org/z/KfYo4nMss))
 
 ### `static_map`
@@ -234,12 +234,19 @@ We plan to add many GPU-accelerated, concurrent data structures to `cuCollection
 #### Examples:
 - [Host-bulk APIs (TODO)]()
 
-### `distinct_count_estimator`
+### `hyperloglog`
 
-`cuco::distinct_count_estimator` implements the well-established [HyperLogLog++ algorithm](https://static.googleusercontent.com/media/research.google.com/de//pubs/archive/40671.pdf) for approximating the count of distinct items in a multiset/stream.
+`cuco::hyperloglog` implements the well-established [HyperLogLog++ algorithm](https://static.googleusercontent.com/media/research.google.com/de//pubs/archive/40671.pdf) for approximating the count of distinct items in a multiset/stream.
 
 #### Examples:
-- [Host-bulk APIs](https://github.com/NVIDIA/cuCollections/blob/dev/examples/distinct_count_estimator/host_bulk_example.cu) (see [live example in godbolt](https://godbolt.org/z/sMfofM6qd))
-- [Device-ref APIs](https://github.com/NVIDIA/cuCollections/blob/dev/examples/distinct_count_estimator/device_ref_example.cu) (see [live example in godbolt](https://godbolt.org/z/156T9ox7h))
+- [Host-bulk APIs](https://github.com/NVIDIA/cuCollections/blob/dev/examples/hyperloglog/host_bulk_example.cu) (see [live example in godbolt](https://godbolt.org/z/G4qdcTezE))
+- [Device-ref APIs](https://github.com/NVIDIA/cuCollections/blob/dev/examples/hyperloglog/device_ref_example.cu) (see [live example in godbolt](https://godbolt.org/z/n88713o4n))
+
+### `bloom_filter`
+
+`cuco::bloom_filter` implements a Blocked Bloom Filter for approximate set membership queries.
+
+#### Examples:
+- [Host-bulk APIs](https://github.com/NVIDIA/cuCollections/blob/dev/examples/bloom_filter/host_bulk_example.cu) (see [live example in godbolt](https://godbolt.org/z/EY7T5v5aE))
 
 

benchmarks/CMakeLists.txt

@@ -35,7 +35,7 @@ function(ConfigureBench BENCH_NAME)
                                         RUNTIME_OUTPUT_DIRECTORY "${CMAKE_BINARY_DIR}/benchmarks")
     target_include_directories(${BENCH_NAME} PRIVATE
                                              "${CMAKE_CURRENT_SOURCE_DIR}")
-    target_compile_options(${BENCH_NAME} PRIVATE --expt-extended-lambda --expt-relaxed-constexpr -lineinfo)
+    target_compile_options(${BENCH_NAME} PRIVATE --expt-extended-lambda -lineinfo)
     target_link_libraries(${BENCH_NAME} PRIVATE
                                         nvbench::main
                                         pthread
@@ -49,53 +49,59 @@ endfunction(ConfigureBench)
 ###################################################################################################
 # - static_set benchmarks -------------------------------------------------------------------------
 ConfigureBench(STATIC_SET_BENCH
-  hash_table/static_set/contains_bench.cu
-  hash_table/static_set/find_bench.cu
-  hash_table/static_set/insert_bench.cu
-  hash_table/static_set/retrieve_all_bench.cu
-  hash_table/static_set/size_bench.cu
-  hash_table/static_set/rehash_bench.cu)
+  static_set/contains_bench.cu
+  static_set/find_bench.cu
+  static_set/insert_bench.cu
+  static_set/retrieve_all_bench.cu
+  static_set/size_bench.cu
+  static_set/rehash_bench.cu)
 
 ###################################################################################################
 # - static_map benchmarks -------------------------------------------------------------------------
 ConfigureBench(STATIC_MAP_BENCH
-  hash_table/static_map/insert_bench.cu
-  hash_table/static_map/find_bench.cu
-  hash_table/static_map/contains_bench.cu
-  hash_table/static_map/erase_bench.cu
-  hash_table/static_map/insert_or_apply_bench.cu)
+  static_map/insert_bench.cu
+  static_map/find_bench.cu
+  static_map/contains_bench.cu
+  static_map/erase_bench.cu
+  static_map/insert_or_apply_bench.cu)
 
 ###################################################################################################
 # - static_multiset benchmarks --------------------------------------------------------------------
 ConfigureBench(STATIC_MULTISET_BENCH
-  hash_table/static_multiset/contains_bench.cu
-  hash_table/static_multiset/retrieve_bench.cu
-  hash_table/static_multiset/count_bench.cu
-  hash_table/static_multiset/find_bench.cu
-  hash_table/static_multiset/insert_bench.cu)
+  static_multiset/contains_bench.cu
+  static_multiset/retrieve_bench.cu
+  static_multiset/count_bench.cu
+  static_multiset/find_bench.cu
+  static_multiset/insert_bench.cu)
 
 ###################################################################################################
 # - static_multimap benchmarks --------------------------------------------------------------------
 ConfigureBench(STATIC_MULTIMAP_BENCH
-  hash_table/static_multimap/insert_bench.cu
-  hash_table/static_multimap/retrieve_bench.cu
-  hash_table/static_multimap/query_bench.cu
-  hash_table/static_multimap/count_bench.cu)
+  static_multimap/insert_bench.cu
+  static_multimap/retrieve_bench.cu
+  static_multimap/query_bench.cu
+  static_multimap/count_bench.cu)
 
 ###################################################################################################
 # - dynamic_map benchmarks ------------------------------------------------------------------------
 ConfigureBench(DYNAMIC_MAP_BENCH
-  hash_table/dynamic_map/insert_bench.cu
-  hash_table/dynamic_map/find_bench.cu
-  hash_table/dynamic_map/contains_bench.cu
-  hash_table/dynamic_map/erase_bench.cu)
+  dynamic_map/insert_bench.cu
+  dynamic_map/find_bench.cu
+  dynamic_map/contains_bench.cu
+  dynamic_map/erase_bench.cu)
 
 ###################################################################################################
 # - hash function benchmarks ----------------------------------------------------------------------
-ConfigureBench(HASH_BENCH
-  hash_bench.cu)
+ConfigureBench(HASH_FUNCTION_BENCH
+  hash_function/hash_function_bench.cu)
 
 ###################################################################################################
-# - distinct_count_estimator benchmarks -----------------------------------------------------------
-ConfigureBench(DISTINCT_COUNT_ESTIMATOR_BENCH
-  distinct_count_estimator_bench.cu)
+# - hyperloglog benchmarks -----------------------------------------------------------
+ConfigureBench(HYPERLOGLOG_BENCH
+  hyperloglog/hyperloglog_bench.cu)
+
+###################################################################################################
+# - bloom_filter benchmarks -----------------------------------------------------------------------
+ConfigureBench(BLOOM_FILTER_BENCH
+  bloom_filter/add_bench.cu
+  bloom_filter/contains_bench.cu)