-
Notifications
You must be signed in to change notification settings - Fork 105
Open
Labels
type: feature requestNew feature requestNew feature request
Description
Part of #110 (Refactor of open address data structures)
Development branch: NVIDIA/cuCollections/refactor
Synopsis
A conceptual list of <key,payload> elements that need not necessarily be contiguous in memory. For example, vector< pair<key,value> > or vector<key> and vector<value> (i.e., AoS vs SoA) could be valid implementations of storage.
TODOs
- Implement classes
cuco::storage::array_of_structsandcuco::storage::struct_of_arraysand (optionally)cuco::storage::array_of_structs_of_arrays(closes [FEA] Add option to select between Struct of Arrays vs Array of Structs #103) - Own the mechanism of atomically updating a given slot using
cuda::atomic_ref(closes [FEA] Migrate fromcuda::atomictocuda::atomic_ref#183) - It should provide some mechanism for loading an immutable "window" of slots. The intention is to provide a standard way of using vector load/store operations
-
Enable querying if concurrent insert/find operations are possibleNo longer desired due to operator mixins
References
- https://godbolt.org/z/9bosKoW4v (distinct "ProbingScheme" and "Storage" concepts)
Metadata
Metadata
Assignees
Labels
type: feature requestNew feature requestNew feature request