Thanks for the great write-up! I have one quibble with your implementation of quadratic probing though: the usual index function used for quadratic probing is start_index + (i + i^2) / 2. This is the sequence you get by…
Have you tried absl::flat_map? It uses simd in a different way than described in this article, and Google claims that it saves them a lot of memory because it still works pretty well at 90-95% occupancy.
Thanks for the great write-up! I have one quibble with your implementation of quadratic probing though: the usual index function used for quadratic probing is start_index + (i + i^2) / 2. This is the sequence you get by…
Have you tried absl::flat_map? It uses simd in a different way than described in this article, and Google claims that it saves them a lot of memory because it still works pretty well at 90-95% occupancy.