While many approximate computing methods are quite application-dependent, reducing the size of the data representation used in the computation has a more general applicability. We present a tuning assistant for floating to fixed point optimization (TAFFO), an LLVM-based framework designed to assist programmers in the precision tuning of software. We discuss the framework architecture and we provide guidelines to effectively tradeoff precision to improve the time-to-solution. We evaluate our framework on a well-known approximate computing benchmark suite, AxBench, achieving a speedup on 5 out of 6 benchmarks (up to 366%) with only a limited loss in precision (<3% for all benchmarks). Contrary to most related tools, TAFFO supports both C and C++ programs. It is provided as a plugin for LLVM, a design solution that improves significantly the maintainability of the tool and its ease of use.
TAFFO: Tuning Assistant for Floating to Fixed point Optimization
Cherubin S.;Cattaneo D.;Chiari M.;Di Bello A.;Agosta G.
2020-01-01
Abstract
While many approximate computing methods are quite application-dependent, reducing the size of the data representation used in the computation has a more general applicability. We present a tuning assistant for floating to fixed point optimization (TAFFO), an LLVM-based framework designed to assist programmers in the precision tuning of software. We discuss the framework architecture and we provide guidelines to effectively tradeoff precision to improve the time-to-solution. We evaluate our framework on a well-known approximate computing benchmark suite, AxBench, achieving a speedup on 5 out of 6 benchmarks (up to 366%) with only a limited loss in precision (<3% for all benchmarks). Contrary to most related tools, TAFFO supports both C and C++ programs. It is provided as a plugin for LLVM, a design solution that improves significantly the maintainability of the tool and its ease of use.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
