Exploration of compiler optimization sequences using clustering-based selection

Abstract

Due to the large number of optimizations provided in modern compilers and to compiler optimization specific opportunities, a Design Space Exploration (DSE) is necessary to search for the best sequence of compiler optimizations for a given code fragment (e.g., function). As this exploration is a complex and time consuming task, in this paper we present DSE strategies to select optimization sequences to both improve the performance of each function and reduce the exploration time. The DSE is based on a clustering approach which groups functions with similarities and then explore the reduced search space provided by the optimizations previously suggested for the functions in each group. The identification of similarities between functions uses a data mining method which is applied to a symbolic code representation of the source code. The DSE process uses the reduced set identified by clustering in two ways: as the design space or as the initial configuration. In both ways, the adoption of a pre-selection based on clustering allows the use of simple and fast DSE algorithms. Our experiments for evaluating the effectiveness of the proposed approach address the exploration of compiler optimization sequences considering 49 compilation passes and targeting a Xilinx MicroBlaze processor, and were performed aiming performance improvements for 41 functions. Experimental results reveal that the use of our new clustering-based DSE approach achieved a significant reduction on the total exploration time of the search space (18× over a Genetic Algorithm approach for DSE) at the same time that important performance speedups (43% over the baseline) were obtained by the optimized codes.

Scalability of Multi-objective Evolutionary Algorithms for Solving Real-World Complex Optimization Problems

“…DAMICORE requires no parameters setup to run (although some execution options may improve its performance). DAMICORE has been successfully employed in a variety of fields, for example, software-hardware co-design [16,17,18], compiler optimization [19,20], student profiling in e-learning environments [21,22], identification of phytopathology from sensor data [18], systematic literature review, identification of cross-cut concerns [23], electrical distribution systems [24], and novel methods in bioinformatics [25]. Feature Sensitivity Analysis.…”

“…The former only works on the objective space for the exclusive purpose of space reduction to determine the essential objective set [9]. Moreover, the FS-OPA preserves the original variable space, which favours non-expert human interpretability (relevant for some classes of real-world problems); it also has a relatively low-time complexity and has shown beneficial results when applied to small datasets [10,[16][17][18][19][20][21][22][23].…”

Machine Learning in Compiler Optimization

“…K-means is also used in the work presented in [93] to summarise the code structures of parallel programs that benefit from similar optimisation strategies. In addition to k-means, Martins et al employed the Fast Newman clustering algorithm [94] which works on network structures to group functions that may benefit from similar compiler optimizations [95].…”

Caracterização do perfil de consumo de recursos de programas binários utilizando a técnica DAMICORE

“…[16] utilizam a ténica para mineracão de dados em ambientes virtuais de Ensino/Aprendizagem. Outros trabalhos abordam a técnica em sistemas reconfiguráveis (FPGA) e compiladores [17,24,14,13].…”