Enhancement of Orchestration Algorithms for Compiler Optimization

Applying the right compiler optimizations to a particular program can have a significant impact on program performance. But achieving this target involves lot of complication because arriving at the compiler configuration for a particular problem is a complex process. The performance of the program measured by time and memory depends on the machine architecture, problem domain and the settings of the compiler.

Due

to the non-linear interaction of compiler optimizations however, determining the best setting is nontrivial. There have been several proposed techniques that search the space of compiler options to find good solutions; however such approaches can be expensive. This paper proposes a different approach using an effective orchestration algorithm strategy. We show that such an approach which outperforms existing orchestration algorithms. In this paper, it is proposed to study the classification of problems, identification of ideal objective functions for different tasks and the ordering of objective function for optimization. In this paper we proposed an automated framework to select the compiler options for a particular problem from large set options. Many previous works consider only limited set of options. For this framework, we implemented compiler optimization selection algorithms such as branch and bound strategy and advanced combined elimination algorithm and evaluated its efficiencies to improve tuning time. Using our technique we achieve better improvement over the highest optimization setting of the MiBench benchmark suite on Intel core duo processor.