1 / 12

Specialized systems are  Inevitable  Already the norm  Practical

Specialized systems are  Inevitable  Already the norm  Practical. Heterogeneous parallelism. Data informatics. Scientific Engineering. Personal Robotics. Virtual Worlds. Programming Heterogeneous Parallelism . Applications. New Arch. Heterogeneous Hardware. 4.

doria
Download Presentation

Specialized systems are  Inevitable  Already the norm  Practical

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Specialized systems are Inevitable Already the norm Practical

  2. Heterogeneous parallelism

  3. Data informatics Scientific Engineering Personal Robotics Virtual Worlds Programming Heterogeneous Parallelism Applications New Arch. Heterogeneous Hardware 4

  4. Architecture Specific Programming Models CUDA OpenCL Threads OpenMP MPI Hadoop Verilog VHDL

  5. High Performance But Low Productivity

  6. Data informatics Scientific Engineering Personal Robotics Virtual Worlds General-Purpose Languages Applications Scala Java Not enough semantic knowledge to compile automatically No restrictions Python C++ Ruby Clojure New Arch. Heterogeneous Hardware

  7. Make Specialized PL Practical Inevitable Already the norm Practical

  8. Domain Specific Languages • Domain Specific Languages (DSLs) • Programming language with restricted expressiveness for a particular domain • High-level, usually declarative, and deterministic Most Science apps start with this PL Most popular parallel PL Enabled GPU innovation

  9. Machine Learning (OptiML) DSL Compiler DSL Compiler DSL Compiler DSL Compiler DSL Compiler Data informatics Scientific Engineering Data Analytics (OptiQL) Graph Alg. (Green Marl) Statistics (R) Physics (Liszt) Personal Robotics Virtual Worlds High Performance Domain-Specific Languages Applications Domain- Specific Languages New Arch. Heterogeneous Hardware

  10. HP DSLs Will Be Practical Embedded DSLs Flexible embedding languages: Scala, F# Metaprogramming (aka staging) Blurring line between libraries and languages DSL Frameworks Support domain specific optimizations Support parallelism and locality optimizations Multiple backends e.g. Spoofax, Rose, Delite, Terra/Lua

More Related