Toward performance portability for CPUS and GPUS through algorithmic compositions

Chang, Li-Wen

Permalink

https://hdl.handle.net/2142/98331 Copy

Description

Title

Toward performance portability for CPUS and GPUS through algorithmic compositions

Author(s)

Chang, Li-Wen

Issue Date

2017-07-05

Director of Research (if dissertation) or Advisor (if thesis)

Hwu, Wen-Mei W.

Doctoral Committee Chair(s)

Hwu, Wen-Mei W.

Committee Member(s)

• Chen, Deming
• Kim, Nam Sung
• Lumetta, Steven S.

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Performance portability
• Algorithmic composition
• Parallel programming
• TANGRAM
• Programming language
• Compiler
• Graphics processing units (GPUs)
• Central processing units (CPUs)
• Open Computing Language (OpenCL)
• Open Multi-Processing (OpenMP)
• Open Accelerators (OpenACC)
• C++ Accelerated Massive Parallelism (C++AMP)

Abstract

The diversity of microarchitecture designs in heterogeneous computing systems allows programs to achieve high performance and energy efficiency, but results in substantial software redevelopment cost for each type or generation of hardware. To mitigate this cost, a performance portable programming system is required. This work presents my solution to the performance portability problem. I argue that a new language is required for replacing the current practices of programming systems to achieve practical performance portability. To support my argument, I first demonstrate the limited performance portability of the current practices by showing quantitative and qualitative evidences. I identify the main limiting issues of conventional programming languages. To overcome the issues, I propose a new modular, composition-based programming language that can effectively express an algorithmic design space with functional polymorphism, and a compiler that can effectively explore the design space and facilitate many high-level optimization techniques. This proposed approach achieves no less than 70% of the performance of highly optimized vendor libraries such as Intel MKL and NVIDIA CUBLAS/CUSPARSE on an Intel i7-3820 Sandy Bridge CPU, an NVIDIA C2050 Fermi GPU, and an NVIDIA K20c Kepler GPU.

Graduation Semester

2017-08

Type of Resource

text

Permalink

http://hdl.handle.net/2142/98331

Copyright and License Information

Copyright 2017 Li-Wen Chang

Owning Collections