Utilizing ensemble learning for performance and power modeling and improvement of parallel cancer deep learning CANDLE benchmarks
Abstract Machine learning (ML) continues to grow in importance across nearly all domains in modeling to learn from data. Often a tradeoff exists between a model's ability to minimize bias and variance. In this article, we utilize ensemble learning to combine linear, nonlinear, and tree‐/rule‐based ML methods to cope with the bias‐variance tradeoff and result in more accurate models. We use the datasets collected for two parallel cancer deep learning CANDLE benchmarks, NT3 and P1B2, to build performance and power models based on hardware performance counters using single‐object and multiple‐objects ensemble learning to identify the most important counters for improvement on the Cray XC40 Theta at Argonne National Laboratory. Based on the insights from these models, we improve the performance and energy of P1B2 and NT3 by optimizing the deep learning environments TensorFlow, Keras, Horovod, and Python under the huge page size of 8 MB. Experimental results show that ensemble learning not only produces more accurate models but also provides more robust performance counter ranking. We achieve up to 61.15% performance improvement and up to 62.58% energy saving for P1B2 and up to 55.81% performance improvement and up to 52.60% energy saving for NT3 on up to 24,576 cores.
- Sponsoring Organization:
- USDOE
- Grant/Contract Number:
- AC02-06CH11357
- OSTI ID:
- 1983428
- Journal Information:
- Concurrency and Computation. Practice and Experience, Journal Name: Concurrency and Computation. Practice and Experience Journal Issue: 15 Vol. 35; ISSN 1532-0626
- Publisher:
- Wiley Blackwell (John Wiley & Sons)Copyright Statement
- Country of Publication:
- United Kingdom
- Language:
- English
Similar Records
Exploring flexible communications for streamlining DNN ensemble training pipelines
Accelerating Collective Communication in Data Parallel Training across Deep Learning Frameworks