Exploiting statistical correlations for proactive prediction of program behaviors

Jiang, Yunlian; Zhang, Eddy Z.; Tian, Kai; Mao, Feng; Gethers, Malcom; Shen, Xipeng; Gao, Yuefang

doi:10.1145/1772954.1772989

Cited by 31 publications

(37 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recently introduced parallel PARSEC benchmark suite [Bienia et al 2008] contains 6 datasets for each benchmark. A large number of datasets is not only useful for compiler research and workload characterization research [Jiang et al 2010] many architecture studies rely on profile-based optimization techniques as well [Magklis et al 2003;Sankaranarayanan and Skadron 2004], and may benefit from having more datasets in order to study dataset sensitivity.…”

Section: Related Workmentioning

confidence: 99%

Deconstructing iterative optimization

Chen¹,

Fang²,

Huang³

et al. 2012

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Iterative optimization is a popular compiler optimization approach that has been studied extensively over the past decade. In this article, we deconstruct iterative optimization by evaluating whether it works across datasets and by analyzing why it works.Up to now, most iterative optimization studies are based on a premise which was never truly evaluated: that it is possible to learn the best compiler optimizations across datasets. In this article, we evaluate this question for the first time with a very large number of datasets. We therefore compose KDataSets, a dataset suite with 1000 datasets for 32 programs, which we release to the public. We characterize the diversity of KDataSets, and subsequently use it to evaluate iterative optimization. For all 32 programs, we find that there exists at least one combination of compiler optimizations that achieves at least 83% or more of the best possible speedup across all datasets on two widely used compilers (Intel's ICC and GNU's GCC). This optimal combination is program-specific and yields speedups up to 3.75× (averaged across datasets of a program) over the highest optimization level of the compilers (-O3 for GCC and -fast for ICC). This finding suggests that optimizing programs across datasets might be much easier than previously anticipated.In addition, we evaluate the idea of introducing compiler choice as part of iterative optimization. We find that it can further improve the performance of iterative optimization because different programs favor different compilers. We also investigate why iterative optimization works by analyzing the optimal combinations. We find that only a handful optimizations yield most of the speedup. Finally, we show that optimizations interact in a complex and sometimes counterintuitive way through two case studies, which confirms that iterative optimization is an irreplaceable and important compiler strategy.

show abstract

Section: Related Workmentioning

confidence: 99%

Deconstructing iterative optimization

Chen¹,

Fang²,

Huang³

et al. 2012

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

show abstract

“…These include static code structures [40] and runtime information such as system load [41] and performance counters [42]. In compiler research, the feature sets used for predictive models are often provided without explanation and rarely is the quality of those features evaluated.…”

Section: Related Workmentioning

confidence: 99%

End-to-End Deep Learning of Optimization Heuristics

Cummins

Petoumenos

Wang

et al. 2017

2017 26th International Conference on Parallel Architectures and Compilation Techniques (PACT)

139

154

View full text Add to dashboard Cite

Accurate automatic optimization heuristics are necessary for dealing with the complexity and diversity of modern hardware and software. Machine learning is a proven technique for learning such heuristics, but its success is bound by the quality of the features used. These features must be hand crafted by developers through a combination of expert domain knowledge and trial and error. This makes the quality of the final model directly dependent on the skill and available time of the system architect.Our work introduces a better way for building heuristics. We develop a deep neural network that learns heuristics over raw code, entirely without using code features. The neural network simultaneously constructs appropriate representations of the code and learns how best to optimize, removing the need for manual feature creation. Further, we show that our neural nets can transfer learning from one optimization problem to another, improving the accuracy of new models, without the help of human experts.We compare the effectiveness of our automatically generated heuristics against ones with features hand-picked by experts. We examine two challenging tasks: predicting optimal mapping for heterogeneous parallelism and GPU thread coarsening factors. In 89% of the cases, the quality of our fully automatic heuristics matches or surpasses that of state-of-theart predictive models using hand-crafted features, providing on average 16% and 12% more performance with no human effort expended on designing features.

show abstract

“…Much research has been devoted to modeling system behavior as a means of prediction for databases [5], [6], cluster computing [30], [31], networking [32], [33], [34], program optimization [35], [36], mapping parallelism [37] etc.…”

Section: Related Workmentioning

confidence: 99%

“…Delving further into extraction of non-trivial features, research has explored extracting predictors from execution traces [41] to model program complexity [8], to improve hardware simulation specificity [42], [43], and to find bugs cooperatively [44]. There has also been research on multi-component systems (e.g., content-distribution networks) where the whole system may not be observable in one place.…”

Section: Related Workmentioning

confidence: 99%

Mantis: Efficient Predictions of Execution Time, Energy Usage, Memory Usage and Network Usage on Smart Mobile Devices

Kwon

Lee

et al. 2015

IEEE Trans. on Mobile Comput.

View full text Add to dashboard Cite

We present Mantis, a framework for predicting the Computational Resource Consumption(CRC) of Android applications on given inputs accurately, and efficiently. A key insight underlying Mantis is that program codes often contain features that correlate with performance and these features can be automatically computed efficiently. Mantis synergistically combines techniques from program analysis and machine learning. It constructs concise CRC models by choosing from many program execution features only a handful that are most correlated with the program's CRC metric yet can be evaluated efficiently from the program's input. We apply program slicing to reduce evaluation time of a feature and automatically generate executable code snippets for efficiently evaluating features. Our evaluation shows that Mantis predicts four CRC metrics of seven Android apps with estimation error in the range of 0-11.1% by executing predictor code spending at most 1.3% of their execution time on Galaxy Nexus.

show abstract

Exploiting statistical correlations for proactive prediction of program behaviors

Cited by 31 publications

References 12 publications

Deconstructing iterative optimization

Deconstructing iterative optimization

End-to-End Deep Learning of Optimization Heuristics

Mantis: Efficient Predictions of Execution Time, Energy Usage, Memory Usage and Network Usage on Smart Mobile Devices

Contact Info

Product

Resources

About