[Outdated] Scaling workspace resources

cpp/bench/ann/src/raft/raft_ann_bench_utils.h

@@ -27,6 +27,7 @@
 #include <rmm/cuda_stream_view.hpp>
 #include <rmm/device_uvector.hpp>
 #include <rmm/mr/device/failure_callback_resource_adaptor.hpp>
+#include <rmm/mr/device/managed_memory_resource.hpp>
 #include <rmm/mr/device/pool_memory_resource.hpp>
 
 #include <memory>
@@ -70,13 +71,14 @@ inline auto rmm_oom_callback(std::size_t bytes, void*) -> bool
  */
 class shared_raft_resources {
  public:
-  using pool_mr_type = rmm::mr::pool_memory_resource<rmm::mr::device_memory_resource>;
-  using mr_type      = rmm::mr::failure_callback_resource_adaptor<pool_mr_type>;
+  using pool_mr_type  = rmm::mr::pool_memory_resource<rmm::mr::device_memory_resource>;
+  using mr_type       = rmm::mr::failure_callback_resource_adaptor<pool_mr_type>;
+  using large_mr_type = rmm::mr::managed_memory_resource;
 
   shared_raft_resources()
   try : orig_resource_{rmm::mr::get_current_device_resource()},
     pool_resource_(orig_resource_, 1024 * 1024 * 1024ull),
-    resource_(&pool_resource_, rmm_oom_callback, nullptr) {
+    resource_(&pool_resource_, rmm_oom_callback, nullptr), large_mr_() {
     rmm::mr::set_current_device_resource(&resource_);
   } catch (const std::exception& e) {
     auto cuda_status = cudaGetLastError();
@@ -99,10 +101,16 @@ class shared_raft_resources {
 
   ~shared_raft_resources() noexcept { rmm::mr::set_current_device_resource(orig_resource_); }
 
+  auto get_large_memory_resource() noexcept
+  {
+    return static_cast<rmm::mr::device_memory_resource*>(&large_mr_);
+  }
+
  private:
   rmm::mr::device_memory_resource* orig_resource_;
   pool_mr_type pool_resource_;
   mr_type resource_;
+  large_mr_type large_mr_;
 };
 
 /**
@@ -123,6 +131,12 @@ class configured_raft_resources {
   explicit configured_raft_resources(const std::shared_ptr<shared_raft_resources>& shared_res)
     : shared_res_{shared_res}, res_{rmm::cuda_stream_view(get_stream_from_global_pool())}
   {
+    // set the large workspace resource to the raft handle, but without the deleter
+    // (this resource is managed by the shared_res).
+    raft::resource::set_large_workspace_resource(
+      res_,
+      std::shared_ptr<rmm::mr::device_memory_resource>(shared_res_->get_large_memory_resource(),
+                                                       raft::void_op{}));
   }
 
   /** Default constructor creates all resources anew. */

cpp/include/raft/core/resource/device_memory_resource.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2022-2023, NVIDIA CORPORATION.
+ * Copyright (c) 2022-2024, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -35,6 +35,16 @@ namespace raft::resource {
  * @{
  */
 
+class memory_resource : public resource {
+ public:
+  explicit memory_resource(std::shared_ptr<rmm::mr::device_memory_resource> mr) : mr_(mr) {}
+  ~memory_resource() override = default;
+  auto get_resource() -> void* override { return mr_.get(); }
+
+ private:
+  std::shared_ptr<rmm::mr::device_memory_resource> mr_;
+};
+
 class limiting_memory_resource : public resource {
  public:
   limiting_memory_resource(std::shared_ptr<rmm::mr::device_memory_resource> mr,
@@ -66,6 +76,29 @@ class limiting_memory_resource : public resource {
   }
 };
 
+/**
+ * Factory that knows how to construct a specific raft::resource to populate
+ * the resources instance.
+ */
+class large_workspace_resource_factory : public resource_factory {
+ public:
+  explicit large_workspace_resource_factory(
+    std::shared_ptr<rmm::mr::device_memory_resource> mr = {nullptr})
+    : mr_{mr ? mr
+             : std::shared_ptr<rmm::mr::device_memory_resource>{
+                 rmm::mr::get_current_device_resource(), void_op{}}}
+  {
+  }
+  auto get_resource_type() -> resource_type override
+  {
+    return resource_type::LARGE_MEMORY_RESOURCE;
+  }
+  auto make_resource() -> resource* override { return new memory_resource(mr_); }
+
+ private:
+  std::shared_ptr<rmm::mr::device_memory_resource> mr_;
+};
+
 /**
  * Factory that knows how to construct a specific raft::resource to populate
  * the resources instance.
@@ -144,7 +177,7 @@ class workspace_resource_factory : public resource_factory {
     // Note, the workspace does not claim all this memory from the start, so it's still usable by
     // the main resource as well.
     // This limit is merely an order for algorithm internals to plan the batching accordingly.
-    return total_size / 2;
+    return total_size / 4;
   }
 };
 
@@ -241,6 +274,21 @@ inline void set_workspace_to_global_resource(
     workspace_resource_factory::default_plain_resource(), allocation_limit, std::nullopt));
 };
 
+inline auto get_large_workspace_resource(resources const& res) -> rmm::mr::device_memory_resource*
+{
+  if (!res.has_resource_factory(resource_type::LARGE_MEMORY_RESOURCE)) {
+    res.add_resource_factory(std::make_shared<large_workspace_resource_factory>());
+  }
+  return res.get_resource<rmm::mr::device_memory_resource>(resource_type::LARGE_MEMORY_RESOURCE);
+};
+
+inline void set_large_workspace_resource(resources const& res,
+                                         std::shared_ptr<rmm::mr::device_memory_resource> mr = {
+                                           nullptr})
+{
+  res.add_resource_factory(std::make_shared<large_workspace_resource_factory>(mr));
+};
+
 /** @} */
 
 }  // namespace raft::resource

cpp/include/raft/core/resource/resource_types.hpp

@@ -42,9 +42,10 @@ enum resource_type {
   STREAM_VIEW,             // view of a cuda stream or a placeholder in
                            // CUDA-free builds
   THRUST_POLICY,           // thrust execution policy
-  WORKSPACE_RESOURCE,      // rmm device memory resource
+  WORKSPACE_RESOURCE,      // rmm device memory resource for small temporary allocations
   CUBLASLT_HANDLE,         // cublasLt handle
   CUSTOM,                  // runtime-shared default-constructible resource
+  LARGE_MEMORY_RESOURCE,   // rmm device memory resource for somewhat large temporary allocations
 
   LAST_KEY  // reserved for the last key
 };

cpp/include/raft/neighbors/detail/cagra/cagra_build.cuh

@@ -97,11 +97,22 @@ void build_knn_graph(raft::resources const& res,
     search_params->lut_dtype = CUDA_R_8U;
     search_params->internal_distance_dtype = CUDA_R_32F;
   }
-  const auto top_k          = node_degree + 1;
-  uint32_t gpu_top_k        = node_degree * refine_rate.value_or(2.0f);
-  gpu_top_k                 = std::min<IdxT>(std::max(gpu_top_k, top_k), dataset.extent(0));
-  const auto num_queries    = dataset.extent(0);
-  const auto max_batch_size = 1024;
+  const auto top_k       = node_degree + 1;
+  uint32_t gpu_top_k     = node_degree * refine_rate.value_or(2.0f);
+  gpu_top_k              = std::min<IdxT>(std::max(gpu_top_k, top_k), dataset.extent(0));
+  const auto num_queries = dataset.extent(0);
+  rmm::mr::device_memory_resource* workspace_mr = raft::resource::get_workspace_resource(res);
+  // Heuristic: how much of the workspace we can spare for the queries.
+  // The rest is going to be used by ivf_pq::search
+  const auto workspace_queries_bytes = raft::resource::get_workspace_free_bytes(res) / 5;
+  auto max_batch_size =
+    std::min<size_t>(workspace_queries_bytes / sizeof(DataT) / dataset.extent(1), 4096);
+  // Heuristic: if the workspace is too small for a decent batch size, switch to use the large
+  // resource with a default batch size.
+  if (max_batch_size < 128) {
+    max_batch_size = 1024;
+    workspace_mr   = raft::resource::get_large_workspace_resource(res);
+  }
   RAFT_LOG_DEBUG(
     "IVF-PQ search node_degree: %d, top_k: %d,  gpu_top_k: %d,  max_batch_size:: %d, n_probes: %u",
     node_degree,
@@ -110,12 +121,17 @@ void build_knn_graph(raft::resources const& res,
     max_batch_size,
     search_params->n_probes);
 
-  auto distances = raft::make_device_matrix<float, int64_t>(res, max_batch_size, gpu_top_k);
-  auto neighbors = raft::make_device_matrix<int64_t, int64_t>(res, max_batch_size, gpu_top_k);
-  auto refined_distances = raft::make_device_matrix<float, int64_t>(res, max_batch_size, top_k);
-  auto refined_neighbors = raft::make_device_matrix<int64_t, int64_t>(res, max_batch_size, top_k);
-  auto neighbors_host    = raft::make_host_matrix<int64_t, int64_t>(max_batch_size, gpu_top_k);
-  auto queries_host = raft::make_host_matrix<DataT, int64_t>(max_batch_size, dataset.extent(1));
+  rmm::mr::device_memory_resource* large_mr = raft::resource::get_large_workspace_resource(res);
+  auto distances                            = raft::make_device_mdarray<float>(
+    res, large_mr, raft::make_extents<int64_t>(max_batch_size, gpu_top_k));
+  auto neighbors = raft::make_device_mdarray<int64_t>(
+    res, large_mr, raft::make_extents<int64_t>(max_batch_size, gpu_top_k));
+  auto refined_distances = raft::make_device_mdarray<float>(
+    res, large_mr, raft::make_extents<int64_t>(max_batch_size, top_k));
+  auto refined_neighbors = raft::make_device_mdarray<int64_t>(
+    res, large_mr, raft::make_extents<int64_t>(max_batch_size, top_k));
+  auto neighbors_host = raft::make_host_matrix<int64_t, int64_t>(max_batch_size, gpu_top_k);
+  auto queries_host   = raft::make_host_matrix<DataT, int64_t>(max_batch_size, dataset.extent(1));
   auto refined_neighbors_host = raft::make_host_matrix<int64_t, int64_t>(max_batch_size, top_k);
   auto refined_distances_host = raft::make_host_matrix<float, int64_t>(max_batch_size, top_k);
 
@@ -124,15 +140,13 @@ void build_knn_graph(raft::resources const& res,
   bool first                    = true;
   const auto start_clock        = std::chrono::system_clock::now();
 
-  rmm::mr::device_memory_resource* device_memory = raft::resource::get_workspace_resource(res);
-
   raft::spatial::knn::detail::utils::batch_load_iterator<DataT> vec_batches(
     dataset.data_handle(),
     dataset.extent(0),
     dataset.extent(1),
     max_batch_size,
     resource::get_cuda_stream(res),
-    device_memory);
+    workspace_mr);
 
   size_t next_report_offset = 0;
   size_t d_report_offset    = dataset.extent(0) / 100;  // Report progress in 1% steps.

cpp/include/raft/neighbors/detail/cagra/graph_core.cuh

@@ -246,15 +246,19 @@ void sort_knn_graph(raft::resources const& res,
   const uint32_t input_graph_degree = knn_graph.extent(1);
   IdxT* const input_graph_ptr       = knn_graph.data_handle();
 
-  auto d_input_graph = raft::make_device_matrix<IdxT, int64_t>(res, graph_size, input_graph_degree);
+  auto large_tmp_mr = resource::get_large_workspace_resource(res);
+
+  auto d_input_graph = raft::make_device_mdarray<IdxT>(
+    res, large_tmp_mr, raft::make_extents<int64_t>(graph_size, input_graph_degree));
 
   //
   // Sorting kNN graph
   //
   const double time_sort_start = cur_time();
   RAFT_LOG_DEBUG("# Sorting kNN Graph on GPUs ");
 
-  auto d_dataset = raft::make_device_matrix<DataT, int64_t>(res, dataset_size, dataset_dim);
+  auto d_dataset = raft::make_device_mdarray<DataT>(
+    res, large_tmp_mr, raft::make_extents<int64_t>(dataset_size, dataset_dim));
   raft::copy(d_dataset.data_handle(),
              dataset_ptr,
              dataset_size * dataset_dim,
@@ -325,6 +329,7 @@ void optimize(raft::resources const& res,
 {
   RAFT_LOG_DEBUG(
     "# Pruning kNN graph (size=%lu, degree=%lu)\n", knn_graph.extent(0), knn_graph.extent(1));
+  auto large_tmp_mr = resource::get_large_workspace_resource(res);
 
   RAFT_EXPECTS(knn_graph.extent(0) == new_graph.extent(0),
                "Each input array is expected to have the same number of rows");
@@ -340,15 +345,16 @@ void optimize(raft::resources const& res,
     //
     // Prune kNN graph
     //
-    auto d_detour_count =
-      raft::make_device_matrix<uint8_t, int64_t>(res, graph_size, input_graph_degree);
+    auto d_detour_count = raft::make_device_mdarray<uint8_t>(
+      res, large_tmp_mr, raft::make_extents<int64_t>(graph_size, input_graph_degree));
 
     RAFT_CUDA_TRY(cudaMemsetAsync(d_detour_count.data_handle(),
                                   0xff,
                                   graph_size * input_graph_degree * sizeof(uint8_t),
                                   resource::get_cuda_stream(res)));
 
-    auto d_num_no_detour_edges = raft::make_device_vector<uint32_t, int64_t>(res, graph_size);
+    auto d_num_no_detour_edges = raft::make_device_mdarray<uint32_t>(
+      res, large_tmp_mr, raft::make_extents<int64_t>(graph_size));
     RAFT_CUDA_TRY(cudaMemsetAsync(d_num_no_detour_edges.data_handle(),
                                   0x00,
                                   graph_size * sizeof(uint32_t),
@@ -468,14 +474,16 @@ void optimize(raft::resources const& res,
                                   graph_size * output_graph_degree * sizeof(IdxT),
                                   resource::get_cuda_stream(res)));
 
-    auto d_rev_graph_count = raft::make_device_vector<uint32_t, int64_t>(res, graph_size);
+    auto d_rev_graph_count = raft::make_device_mdarray<uint32_t>(
+      res, large_tmp_mr, raft::make_extents<int64_t>(graph_size));
     RAFT_CUDA_TRY(cudaMemsetAsync(d_rev_graph_count.data_handle(),
                                   0x00,
                                   graph_size * sizeof(uint32_t),
                                   resource::get_cuda_stream(res)));
 
-    auto dest_nodes   = raft::make_host_vector<IdxT, int64_t>(graph_size);
-    auto d_dest_nodes = raft::make_device_vector<IdxT, int64_t>(res, graph_size);
+    auto dest_nodes = raft::make_host_vector<IdxT, int64_t>(graph_size);
+    auto d_dest_nodes =
+      raft::make_device_mdarray<IdxT>(res, large_tmp_mr, raft::make_extents<int64_t>(graph_size));
 
     for (uint64_t k = 0; k < output_graph_degree; k++) {
 #pragma omp parallel for

cpp/include/raft/neighbors/detail/cagra/utils.hpp

@@ -184,8 +184,11 @@ class device_matrix_view_from_host {
     device_ptr = reinterpret_cast<T*>(attr.devicePointer);
     if (device_ptr == NULL) {
       // allocate memory and copy over
-      device_mem_.emplace(
-        raft::make_device_matrix<T, IdxT>(res, host_view.extent(0), host_view.extent(1)));
+      // NB: We use the temporary "large" workspace resource here; this structure is supposed to
+      // live on stack and not returned to a user.
+      // The user may opt to set this resource to managed memory to allow large allocations.
+      device_mem_.emplace(make_device_mdarray<T, IdxT>(
+        res, resource::get_large_workspace_resource(res), host_view.extents()));
       raft::copy(device_mem_->data_handle(),
                  host_view.data_handle(),
                  host_view.extent(0) * host_view.extent(1),

cpp/include/raft/neighbors/detail/ivf_flat_build.cuh

@@ -22,6 +22,7 @@
 #include <raft/core/nvtx.hpp>
 #include <raft/core/operators.hpp>
 #include <raft/core/resource/cuda_stream.hpp>
+#include <raft/core/resource/device_memory_resource.hpp>
 #include <raft/core/resources.hpp>
 #include <raft/linalg/add.cuh>
 #include <raft/linalg/map.cuh>
@@ -180,7 +181,8 @@ void extend(raft::resources const& handle,
   RAFT_EXPECTS(new_indices != nullptr || index->size() == 0,
                "You must pass data indices when the index is non-empty.");
 
-  auto new_labels = raft::make_device_vector<LabelT, IdxT>(handle, n_rows);
+  auto new_labels = raft::make_device_mdarray<LabelT>(
+    handle, resource::get_large_workspace_resource(handle), raft::make_extents<IdxT>(n_rows));
   raft::cluster::kmeans_balanced_params kmeans_params;
   kmeans_params.metric = index->metric();
   auto orig_centroids_view =
@@ -211,7 +213,8 @@ void extend(raft::resources const& handle,
   }
 
   auto* list_sizes_ptr    = index->list_sizes().data_handle();
-  auto old_list_sizes_dev = raft::make_device_vector<uint32_t, IdxT>(handle, n_lists);
+  auto old_list_sizes_dev = raft::make_device_mdarray<uint32_t>(
+    handle, resource::get_workspace_resource(handle), raft::make_extents<IdxT>(n_lists));
   copy(old_list_sizes_dev.data_handle(), list_sizes_ptr, n_lists, stream);
 
   // Calculate the centers and sizes on the new data, starting from the original values
@@ -367,7 +370,8 @@ inline auto build(raft::resources const& handle,
     auto trainset_ratio = std::max<size_t>(
       1, n_rows / std::max<size_t>(params.kmeans_trainset_fraction * n_rows, index.n_lists()));
     auto n_rows_train = n_rows / trainset_ratio;
-    rmm::device_uvector<T> trainset(n_rows_train * index.dim(), stream);
+    rmm::device_uvector<T> trainset(
+      n_rows_train * index.dim(), stream, raft::resource::get_large_workspace_resource(handle));
     // TODO: a proper sampling
     RAFT_CUDA_TRY(cudaMemcpy2DAsync(trainset.data(),
                                     sizeof(T) * index.dim(),
@@ -427,7 +431,8 @@ inline void fill_refinement_index(raft::resources const& handle,
   common::nvtx::range<common::nvtx::domain::raft> fun_scope(
     "ivf_flat::fill_refinement_index(%zu, %u)", size_t(n_queries));
 
-  rmm::device_uvector<LabelT> new_labels(n_queries * n_candidates, stream);
+  rmm::device_uvector<LabelT> new_labels(
+    n_queries * n_candidates, stream, raft::resource::get_workspace_resource(handle));
   auto new_labels_view =
     raft::make_device_vector_view<LabelT, IdxT>(new_labels.data(), n_queries * n_candidates);
   linalg::map_offset(