Industrial Data Science
in C# and .NET:
Simple. Fast. Reliable.
 
 

ILNumerics - Technical Computing

Modern High Performance Tools for Technical

Computing and Visualization in Industry and Science

Accelerator roadmap 

This page is under construction. Following, additional functionality is planned for ILNumerics Accelerator: 

  • support more functions / function categories for low-level kernel building: 
    • low-level / elementary functions: 
      • complex & fcomplex element types
      • boolean functions: any, all, lt / <, eq / ==, ... 
      • type conversion functions
      • ensure exact same segment semantic as for ILMath / numpy handling of NaN, keeping of reduced dims, empty / scalar arrays, ...
    • higher-level functions:
      • linear algebra (lapack), fft
      • implement toolbox functions 
  • enable more auto-optimizations: 
    • automatic (in)dependency detection in loops
    • automatic loop transforms
    • better binary broadcasting segment switches
    • even more useful kernel code specializations
  • enable / improve automatic mode
    • improve cross device scheduling efficiency 
    • make recovering from buggy OpenCL GPU drivers more robust
  • improve latency hiding 
    • file caching of kernels