Performance Modeling using a Genetic Programming Based Model Error Correction Procedure

Performance modeling;Genetic programming;Error correction procedure

Raghavachar, Kavitha ; Dr.Ranji S Ranjithan, Committee Co-Chair,Dr.John W Baugh, Committee Co-Chair,Dr.G Mahinthakumar, Committee Chair,Raghavachar, Kavitha ; Dr.Ranji S Ranjithan ; Committee Co-Chair ; Dr.John W Baugh ; Committee Co-Chair ; Dr.G Mahinthakumar ; Committee Chair

University:North Carolina State University

关键词: Performance modeling;    Genetic programming;    Error correction procedure;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/1157/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

Application performance models provide insight to designers of high performance computing (HPC) systems on the role of subsystems such as the processor or the network in determining application performance and allow HPC centers to more accurately target procurements to resource requirements. Performance models can also be used to identify application performance bottlenecks and to provide insights about scalability issues. The suitability of a performance model, however, for a particular performance investigation is a function of both the accuracy and the cost of the model. A semi-empirical model developed in an earlier publication for an astrophysics application was shown to be inaccurate when predicting communication cost for large numbers of processors. It was hypothesized that this deficiency is due to the inability of the model to adequately capture communication contention (threshold effects) as well as other un-modeled components such as noise and I⁄O contention. This thesis demonstrates a new approach to capture these unknown features to improve the predictive capabilities of the model. This approach uses a systematic model error correction procedure that uses evolutionary algorithms to find an error correction term to augment the existing model. Four variations of this procedure were investigated and all were shown to produce improved results than the old model. Successful cross-platform application of this approach showed that it adequately captures machine dependent characteristics.This approach was then extended to a second application, which too showed improved results than the standard semi-empirical modeling approach.