reference_database.cuh File Reference

#include <cuda_runtime.h>
#include <cub/device/device_radix_sort.cuh>
#include <cub/device/device_scan.cuh>
#include <cub/device/device_segmented_sort.cuh>
#include <cub/device/device_select.cuh>
#include <cub/device/device_transform.cuh>
#include <cuda/algorithm>
#include <cuda/buffer>
#include <cuda/iterator>
#include <cuda/memory_pool>
#include <cuda/std/cstdint>
#include <cuda/std/functional>
#include <cuda/std/optional>
#include <cuda/std/utility>
#include <cuda/stream>
#include <algorithm>
#include <array>
#include <atomic>
#include <bit>
#include <cstddef>
#include <functional>
#include <limits>
#include <map>
#include <memory>
#include <mutex>
#include <optional>
#include <span>
#include <string>
#include <type_traits>
#include <utility>
#include <vector>
#include <cuddl/detail/hash.cuh>
#include <cuddl/detail/kernels.cuh>
#include <cuddl/detail/pinned_pool.cuh>
#include <cuddl/detail/reference_deltas.cuh>
#include <cuddl/detail/register.cuh>
#include <cuddl/device_span.cuh>
#include <cuddl/error.hpp>
#include <cuddl/pairwise_counts.cuh>
#include <cuddl/reference_index.cuh>

Go to the source code of this file.

Classes
struct	cuddl::score_compatibility
	Construction parameters of compatible score rows. More...
struct	cuddl::reference_database_metadata
	Recorded metadata for an immutable reference database. More...
struct	cuddl::reference_search_result
	Stable reference ID and exact query-relative bucket comparison counts. More...
struct	cuddl::batch_search_result
	Stable query/reference IDs and their exact query-relative bucket comparison counts. More...
struct	cuddl::batch_result_tile
	Non-owning view of one batch-search tile's results. More...
class	cuddl::host_batch_results
	Host copy of a batch_result_tile holding only the passing results, packed in query-major, reference order, in pinned memory; tile views it. More...
struct	cuddl::batch_search_requirements
	Caller-owned storage requirements for one bounded query tile. More...
struct	cuddl::search_options
	Matching-bucket threshold, independent of whether an index accelerates the search. More...
struct	cuddl::all_to_all_shard
	Part cuDDL of count disjoint parts of an all-to-all search. More...
struct	cuddl::detail::sparse_index_key
struct	cuddl::detail::sparse_reference_id
struct	cuddl::detail::sparse_segment_offset
class	cuddl::all_to_all_tile_queue
	Thread-safe source of the all-to-all tiles one process searches. More...
struct	cuddl::all_to_all_tile_queue::tile
class	cuddl::detail::reference_database_view< K, BucketCount, Layout >
	Non-owning, trivially copyable view of one selected reference-row backing. More...
class	cuddl::reference_database< K, BucketCount, Layout >
	Move-only owner of one immutable contiguous reference database. More...

Namespaces
namespace	cuddl
	Batch operations over contiguous device rows.
namespace	cuddl::detail

Typedefs
using	cuddl::detail::device_scope = ::cuda::__ensure_current_context
	Makes stream's device current for the enclosing scope and restores the previous context on exit.

Enumerations
enum class	cuddl::index_storage { cuddl::automatic , cuddl::dense , cuddl::sparse }
	Dense offsets favor query latency; sparse sorted keys reduce index memory and build time. More...

Functions
Result< void >	cuddl::download (batch_result_tile const &tile, cuda::stream_ref stream, host_batch_results &host)
	Copies a device tile's passing results into host's pinned buffers, growing them only when needed, and synchronizes stream.
Result< host_batch_results >	cuddl::download (batch_result_tile const &tile, cuda::stream_ref stream)
	Copies a device tile's passing results to new pinned host memory; see the overload that downloads into reusable host_batch_results.
template<typename F >
Result< void >	cuddl::for_each_passing (batch_result_tile const &tile, cuda::stream_ref stream, F &&f)
	Calls f on the host with every passing result of a device tile, query-major and in reference order; synchronizes stream.
template<uint32_t K, size_t BucketCount, typename Layout = default_register_layout>
Result< void >	cuddl::detail::validate_score_compatibility (score_compatibility const &compatibility)
template<uint32_t K, size_t BucketCount, typename Layout = default_register_layout>
Result< void >	cuddl::detail::validate_non_indexed_score_compatibility (score_compatibility const &compatibility)
template<uint32_t K, size_t BucketCount, typename Layout = default_register_layout>
Result< void >	cuddl::detail::validate_indexed_score_compatibility (score_compatibility const &compatibility)
constexpr uint64_t	cuddl::detail::indexed_cell_count (score_compatibility const &compatibility) noexcept
constexpr uint64_t	cuddl::detail::indexed_posting_count (uint32_t reference_count, score_compatibility const &compatibility) noexcept
constexpr index_storage	cuddl::detail::smaller_index_storage (uint32_t reference_count, score_compatibility const &compatibility) noexcept
	Storage needing fewer device bytes for reference_count references; dense on ties.
uintptr_t	cuddl::detail::align_up (uintptr_t address, size_t alignment) noexcept
constexpr uint32_t	cuddl::detail::batch_query_tile_size (uint32_t reference_count, uint32_t query_count, uint32_t tile_limit=batch_query_tile_count) noexcept
constexpr uint32_t	cuddl::detail::all_to_all_tile_queries (uint32_t reference_count, uint32_t parts=1U) noexcept
	Query rows one all-to-all tile compares: about 2^25 pairs, and at least batch_query_tile_count.
uint32_t	cuddl::detail::external_batch_query_limit (uint32_t reference_count, cuda::device_ref device)
	Query rows one external batch tile may hold on device.

Variables
constexpr double	cuddl::detail::index_pair_fraction_limit = 0.03
	Index pair fraction (reference_index::pair_fraction) above which batch search compares bit-planes for every pair instead of counting through the index.
constexpr uint32_t	cuddl::detail::batch_query_tile_count = 128U
	Minimum query rows per batch tile.