Generate dataset of select_k times (#1497)

This adds an optional flag (`--select_k_dataset`) to the MATRIX_BENCH that will turn on generating a grid search of benchmarks for different select_k algorithms.  Since this adds about 100x as many benchmarks to run as previous (90k vs 900), this is opt-in only right now. This will be used to learn a heuristic function in #1455

This also integrates the faiss block select top-k algorithm into this benchmarking, so that we can compare how it performs against the other select_k algorithms

Authors:
  - Ben Frederickson (https://github.com/benfred)

Approvers:
  - Corey J. Nolet (https://github.com/cjnolet)

URL: #1497

cpp/CMakeLists.txt

@@ -323,6 +323,10 @@ if(RAFT_COMPILE_LIBRARY)
     src/neighbors/detail/selection_faiss_size_t_double.cu
     src/neighbors/detail/selection_faiss_size_t_float.cu
     src/neighbors/detail/selection_faiss_uint32_t_float.cu
+    src/neighbors/detail/selection_faiss_int64_t_double.cu
+    src/neighbors/detail/selection_faiss_int64_t_half.cu
+    src/neighbors/detail/selection_faiss_uint32_t_double.cu
+    src/neighbors/detail/selection_faiss_uint32_t_half.cu
     src/neighbors/ivf_flat_build_float_int64_t.cu
     src/neighbors/ivf_flat_build_int8_t_int64_t.cu
     src/neighbors/ivf_flat_build_uint8_t_int64_t.cu

cpp/bench/prims/CMakeLists.txt

@@ -116,8 +116,16 @@ if(BUILD_PRIMS_BENCH)
   )
 
   ConfigureBench(
-    NAME MATRIX_BENCH PATH bench/prims/matrix/argmin.cu bench/prims/matrix/gather.cu
-    bench/prims/matrix/select_k.cu bench/prims/main.cpp OPTIONAL LIB EXPLICIT_INSTANTIATE_ONLY
+    NAME
+    MATRIX_BENCH
+    PATH
+    bench/prims/matrix/argmin.cu
+    bench/prims/matrix/gather.cu
+    bench/prims/matrix/select_k.cu
+    bench/prims/matrix/main.cpp
+    OPTIONAL
+    LIB
+    EXPLICIT_INSTANTIATE_ONLY
   )
 
   ConfigureBench(

cpp/bench/prims/matrix/main.cpp

@@ -0,0 +1,41 @@
+/*
+ * Copyright (c) 2023, NVIDIA CORPORATION.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include <benchmark/benchmark.h>
+#include <cstring>
+
+namespace raft::matrix {
+void add_select_k_dataset_benchmarks();
+}
+
+int main(int argc, char** argv)
+{
+  // if we're passed a 'select_k_dataset' flag, add in extra benchmarks
+  for (int i = 1; i < argc; ++i) {
+    if (strcmp(argv[i], "--select_k_dataset") == 0) {
+      raft::matrix::add_select_k_dataset_benchmarks();
+
+      // pop off the cmdline argument from argc/argv
+      for (int j = i; j < argc - 1; ++j)
+        argv[j] = argv[j + 1];
+      argc--;
+      break;
+    }
+  }
+  benchmark::Initialize(&argc, argv);
+  if (::benchmark::ReportUnrecognizedArguments(argc, argv)) return 1;
+  benchmark::RunSpecifiedBenchmarks();
+}

cpp/bench/prims/matrix/select_k.cu

@@ -36,7 +36,6 @@
 #include <type_traits>
 
 namespace raft::matrix {
-
 using namespace raft::bench;  // NOLINT
 
 template <typename KeyT, typename IdxT, select::Algo Algo>
@@ -72,17 +71,16 @@ struct selection : public fixture {
 
   void run_benchmark(::benchmark::State& state) override  // NOLINT
   {
-    device_resources handle{stream};
     try {
       std::ostringstream label_stream;
       label_stream << params_.batch_size << "#" << params_.len << "#" << params_.k;
       if (params_.use_same_leading_bits) { label_stream << "#same-leading-bits"; }
       state.SetLabel(label_stream.str());
-      loop_on_state(state, [this, &handle]() {
+      loop_on_state(state, [this]() {
         select::select_k_impl<KeyT, IdxT>(handle,
                                           Algo,
                                           in_dists_.data(),
-                                          in_ids_.data(),
+                                          params_.use_index_input ? in_ids_.data() : NULL,
                                           params_.batch_size,
                                           params_.len,
                                           params_.k,
@@ -182,4 +180,91 @@ SELECTION_REGISTER(double, int64_t, kWarpFiltered);           // NOLINT
 SELECTION_REGISTER(double, int64_t, kWarpDistributed);        // NOLINT
 SELECTION_REGISTER(double, int64_t, kWarpDistributedShm);     // NOLINT
 
+// For learning a heuristic of which selection algorithm to use, we
+// have a couple of additional constraints when generating the dataset:
+// 1. We want these benchmarks to be optionally enabled from the commandline -
+//  there are thousands of them, and the run-time is non-trivial. This should be opt-in only
+// 2. We test out larger k values - that won't work for all algorithms. This requires filtering
+// the input parameters per algorithm.
+// This makes the code to generate this dataset different from the code above to
+// register other benchmarks
+#define SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, A, input)                               \
+  {                                                                                       \
+    using SelectK = selection<KeyT, IdxT, select::Algo::A>;                               \
+    std::stringstream name;                                                               \
+    name << "SelectKDataset/" << #KeyT "/" #IdxT "/" #A << "/" << input.batch_size << "/" \
+         << input.len << "/" << input.k << "/" << input.use_index_input;                  \
+    auto* b = ::benchmark::internal::RegisterBenchmarkInternal(                           \
+      new raft::bench::internal::Fixture<SelectK, select::params>(name.str(), input));    \
+    b->UseManualTime();                                                                   \
+    b->Unit(benchmark::kMillisecond);                                                     \
+  }
+
+const static size_t MAX_MEMORY = 16 * 1024 * 1024 * 1024ULL;
+
+// registers the input for all algorithms
+#define SELECTION_REGISTER_INPUT(KeyT, IdxT, input)                            \
+  {                                                                            \
+    size_t mem = input.batch_size * input.len * (sizeof(KeyT) + sizeof(IdxT)); \
+    if (mem < MAX_MEMORY) {                                                    \
+      SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kRadix8bits, input)            \
+      SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kRadix11bits, input)           \
+      SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kRadix11bitsExtraPass, input)  \
+      if (input.k <= raft::matrix::detail::select::warpsort::kMaxCapacity) {   \
+        SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kWarpImmediate, input)       \
+        SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kWarpFiltered, input)        \
+        SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kWarpDistributed, input)     \
+        SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kWarpDistributedShm, input)  \
+      }                                                                        \
+      if (input.k <= raft::neighbors::detail::kFaissMaxK<IdxT, KeyT>()) {      \
+        SELECTION_REGISTER_ALGO_INPUT(KeyT, IdxT, kFaissBlockSelect, input)    \
+      }                                                                        \
+    }                                                                          \
+  }
+
+void add_select_k_dataset_benchmarks()
+{
+  // define a uniform grid
+  std::vector<select::params> inputs;
+
+  size_t grid_increment = 1;
+  std::vector<int> k_vals;
+  for (size_t k = 0; k < 13; k += grid_increment) {
+    k_vals.push_back(1 << k);
+  }
+  // Add in values just past the limit for warp/faiss select
+  k_vals.push_back(257);
+  k_vals.push_back(2049);
+
+  const static bool select_min = true;
+  const static bool use_ids    = false;
+
+  for (size_t row = 0; row < 13; row += grid_increment) {
+    for (size_t col = 10; col < 28; col += grid_increment) {
+      for (auto k : k_vals) {
+        inputs.push_back(
+          select::params{size_t(1 << row), size_t(1 << col), k, select_min, use_ids});
+      }
+    }
+  }
+
+  // also add in some random values
+  std::default_random_engine rng(42);
+  std::uniform_real_distribution<> row_dist(0, 13);
+  std::uniform_real_distribution<> col_dist(10, 28);
+  std::uniform_real_distribution<> k_dist(0, 13);
+  for (size_t i = 0; i < 1024; ++i) {
+    auto row = static_cast<size_t>(pow(2, row_dist(rng)));
+    auto col = static_cast<size_t>(pow(2, col_dist(rng)));
+    auto k   = static_cast<int>(pow(2, k_dist(rng)));
+    inputs.push_back(select::params{row, col, k, select_min, use_ids});
+  }
+
+  for (auto& input : inputs) {
+    SELECTION_REGISTER_INPUT(double, int64_t, input);
+    SELECTION_REGISTER_INPUT(double, uint32_t, input);
+    SELECTION_REGISTER_INPUT(float, int64_t, input);
+    SELECTION_REGISTER_INPUT(float, uint32_t, input);
+  }
+}
 }  // namespace raft::matrix

cpp/include/raft/neighbors/detail/selection_faiss-ext.cuh

@@ -18,6 +18,7 @@
 
 #include <cstddef>                                            // size_t
 #include <cstdint>                                            // uint32_t
+#include <cuda_fp16.h>                                        // __half
 #include <raft/neighbors/detail/selection_faiss_helpers.cuh>  // kFaissMaxK
 #include <raft/util/raft_explicit.hpp>                        // RAFT_EXPLICIT
 
@@ -58,4 +59,9 @@ instantiate_raft_neighbors_detail_select_k(size_t, double);
 instantiate_raft_neighbors_detail_select_k(int, double);
 instantiate_raft_neighbors_detail_select_k(size_t, float);
 
+instantiate_raft_neighbors_detail_select_k(uint32_t, double);
+instantiate_raft_neighbors_detail_select_k(int64_t, double);
+instantiate_raft_neighbors_detail_select_k(uint32_t, __half);
+instantiate_raft_neighbors_detail_select_k(int64_t, __half);
+
 #undef instantiate_raft_neighbors_detail_select_k

cpp/internal/raft_internal/matrix/select_k.cuh

@@ -20,6 +20,7 @@
 #include <raft/matrix/detail/select_radix.cuh>
 #include <raft/matrix/detail/select_warpsort.cuh>
 #include <raft/matrix/select_k.cuh>
+#include <raft/neighbors/detail/selection_faiss.cuh>
 
 namespace raft::matrix::select {
 
@@ -52,7 +53,8 @@ enum class Algo {
   kWarpImmediate,
   kWarpFiltered,
   kWarpDistributed,
-  kWarpDistributedShm
+  kWarpDistributedShm,
+  kFaissBlockSelect
 };
 
 inline auto operator<<(std::ostream& os, const Algo& algo) -> std::ostream&
@@ -67,6 +69,7 @@ inline auto operator<<(std::ostream& os, const Algo& algo) -> std::ostream&
     case Algo::kWarpFiltered: return os << "kWarpFiltered";
     case Algo::kWarpDistributed: return os << "kWarpDistributed";
     case Algo::kWarpDistributedShm: return os << "kWarpDistributedShm";
+    case Algo::kFaissBlockSelect: return os << "kFaissBlockSelect";
     default: return os << "unknown enum value";
   }
 }
@@ -154,7 +157,9 @@ void select_k_impl(const device_resources& handle,
       return detail::select::warpsort::
         select_k_impl<T, IdxT, detail::select::warpsort::warp_sort_distributed_ext>(
           in, in_idx, batch_size, len, k, out, out_idx, select_min, stream);
+    case Algo::kFaissBlockSelect:
+      return neighbors::detail::select_k(
+        in, in_idx, batch_size, len, out, out_idx, select_min, k, stream);
   }
 }
-
 }  // namespace raft::matrix::select

cpp/src/neighbors/detail/selection_faiss_00_generate.py

@@ -57,6 +57,10 @@
 
 types = dict(
     uint32_t_float=("uint32_t", "float"),
+    uint32_t_double=("uint32_t", "double"),
+    uint32_t_half=("uint32_t", "half"),
+    int64_t_double=("int64_t", "double"),
+    int64_t_half=("int64_t", "half"),
     int32_t_float=("int32_t", "float"),
     long_float=("long", "float"),
     size_t_double=("size_t", "double"),

cpp/src/neighbors/detail/selection_faiss_int64_t_double.cu

@@ -0,0 +1,44 @@
+
+/*
+ * Copyright (c) 2023, NVIDIA CORPORATION.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/*
+ * NOTE: this file is generated by selection_faiss_00_generate.py
+ *
+ * Make changes there and run in this directory:
+ *
+ * > python selection_faiss_00_generate.py
+ *
+ */
+
+#include <cstddef>  // size_t
+#include <cstdint>  // uint32_t
+#include <raft/neighbors/detail/selection_faiss-inl.cuh>
+
+#define instantiate_raft_neighbors_detail_select_k(payload_t, key_t)    \
+  template void raft::neighbors::detail::select_k(const key_t* inK,     \
+                                                  const payload_t* inV, \
+                                                  size_t n_rows,        \
+                                                  size_t n_cols,        \
+                                                  key_t* outK,          \
+                                                  payload_t* outV,      \
+                                                  bool select_min,      \
+                                                  int k,                \
+                                                  cudaStream_t stream)
+
+instantiate_raft_neighbors_detail_select_k(int64_t, double);
+
+#undef instantiate_raft_neighbors_detail_select_k

cpp/src/neighbors/detail/selection_faiss_int64_t_half.cu

@@ -0,0 +1,44 @@
+
+/*
+ * Copyright (c) 2023, NVIDIA CORPORATION.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/*
+ * NOTE: this file is generated by selection_faiss_00_generate.py
+ *
+ * Make changes there and run in this directory:
+ *
+ * > python selection_faiss_00_generate.py
+ *
+ */
+
+#include <cstddef>  // size_t
+#include <cstdint>  // uint32_t
+#include <raft/neighbors/detail/selection_faiss-inl.cuh>
+
+#define instantiate_raft_neighbors_detail_select_k(payload_t, key_t)    \
+  template void raft::neighbors::detail::select_k(const key_t* inK,     \
+                                                  const payload_t* inV, \
+                                                  size_t n_rows,        \
+                                                  size_t n_cols,        \
+                                                  key_t* outK,          \
+                                                  payload_t* outV,      \
+                                                  bool select_min,      \
+                                                  int k,                \
+                                                  cudaStream_t stream)
+
+instantiate_raft_neighbors_detail_select_k(int64_t, half);
+
+#undef instantiate_raft_neighbors_detail_select_k

cpp/src/neighbors/detail/selection_faiss_uint32_t_double.cu

@@ -0,0 +1,44 @@
+
+/*
+ * Copyright (c) 2023, NVIDIA CORPORATION.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/*
+ * NOTE: this file is generated by selection_faiss_00_generate.py
+ *
+ * Make changes there and run in this directory:
+ *
+ * > python selection_faiss_00_generate.py
+ *
+ */
+
+#include <cstddef>  // size_t
+#include <cstdint>  // uint32_t
+#include <raft/neighbors/detail/selection_faiss-inl.cuh>
+
+#define instantiate_raft_neighbors_detail_select_k(payload_t, key_t)    \
+  template void raft::neighbors::detail::select_k(const key_t* inK,     \
+                                                  const payload_t* inV, \
+                                                  size_t n_rows,        \
+                                                  size_t n_cols,        \
+                                                  key_t* outK,          \
+                                                  payload_t* outV,      \
+                                                  bool select_min,      \
+                                                  int k,                \
+                                                  cudaStream_t stream)
+
+instantiate_raft_neighbors_detail_select_k(uint32_t, double);
+
+#undef instantiate_raft_neighbors_detail_select_k