All Stories

  1. Shray: An Owner-Compute Distributed Shared-Memory System
  2. Rank-Polymorphism for Shape-Guided Blocking
  3. Modulo in high-performance code: strength reduction for modulo-based array indexing in loops
  4. Type Patterns: Pattern Matching on the Shapes of Multi-Dimensional Array Types
  5. On Generating Out-Of-Core GPU Code for Multi-Dimensional Array Operations
  6. Parallel scan as a multidimensional array problem
  7. On Mapping N-Dimensional Data-Parallelism Efficiently into GPU-Thread-Spaces
  8. Improving Code Generation for Reductions on Large Nested Data Structures.
  9. Array languages make neural networks fast
  10. Effective Host-GPU Memory Management Through Code Generation