Caching - 5.2 English - 68552

AOCL API Guide (68552)

Document ID
68552
Release Date
2025-12-29
Version
5.2 English

Our implementation uses a kernel cache to accelerate computations by storing previously calculated kernel matrix values. This caching mechanism significantly reduces computational overhead, especially for larger datasets where the same kernel values would otherwise be repeatedly calculated.

Note

You can control the kernel cache size with the cache size option (default: 200 MB). When the cache reaches capacity, a least recently used (LRU) strategy evicts older entries.