Recurrent Neural Network for Classifying of Hpc Applications

Pham, Trong-Ton; Pister, Mathieu; Couvée, Philippe

doi:10.23919/springsim.2019.8732923

Cited by 5 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast to white-box (i.e., first-principles) performance models, accurate black-box approaches for describing the performance and scalability of highly parallel programs have been available for some time. These typically employ curve fitting, machine learning, and general AI methods [28], [29]. A lot of related research has also been done on code optimization, focusing on new data structures, efficient algorithms, and parallelization techniques, all of which require explicit programming [30].…”

Section: Performance Modeling and Optimizationmentioning

confidence: 99%

The Role of Idle Waves, Desynchronization, and Bottleneck Evasion in the Performance of Parallel Programs

Afzal¹,

Hager²,

Wellein³

2022

Preprint

View full text Add to dashboard Cite

The performance of highly parallel applications on distributed-memory systems is influenced by many factors. Analytic performance modeling techniques aim to provide insight into performance limitations and are often the starting point of optimization efforts. However, coupling analytic models across the system hierarchy (socket, node, network) fails to encompass the intricate interplay between the program code and the hardware, especially when execution and communication bottlenecks are involved. In this paper we investigate the effect of bottleneck evasion and how it can lead to automatic overlap of communication overhead with computation. Bottleneck evasion leads to a gradual loss of the initial bulk-synchronous behavior of a parallel code so that its processes become desynchronized. This occurs most prominently in memory-bound programs, which is why we choose memory-bound benchmark and application codes, specifically an MPI-augmented STREAM Triad, sparse matrix-vector multiplication, and a collective-avoiding Chebyshev filter diagonalization code to demonstrate the consequences of desynchronization on two different supercomputing platforms. We investigate the role of idle waves as possible triggers for desynchronization and show the impact of automatic asynchronous communication for a spectrum of code properties and parameters, such as saturation point, matrix structures, domain decomposition, and communication concurrency. Our findings reveal how eliminating synchronization points (such as collective communication or barriers) precipitates performance improvements that go beyond what can be expected by simply subtracting the overhead of the collective from the overall runtime.

show abstract

Section: Performance Modeling and Optimizationmentioning

confidence: 99%

The Role of Idle Waves, Desynchronization, and Bottleneck Evasion in the Performance of Parallel Programs

Afzal¹,

Hager²,

Wellein³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Using the Fixed Window Split technique [16], we divided long sequences into separate sequences after a fixed window size of N timesteps. For instance, if one sequence has a length of 80, it will be divided into 8 sub-sequences with a split size of N =10.…”

Section: E On Context Length and Its Influencementioning

confidence: 99%

Towards Sequential Multivariate Fault Prediction for Vehicular Predictive Maintenance

Hafeez

Alonso

Ter-Sarkisov

2021

2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA)

View full text Add to dashboard Cite

Predictive maintenance, which has traditionally used anomaly detection methods on sensory data, is now being replaced by event-based techniques. These methods utilise events with multiple temporal features, produced by diagnostic modules. This raises the need for predicting the next fault event in industrial machines, specially vehicles, that use Diagnostic Trouble Codes (DTCs). We propose a predictive maintenance approach, named Sequential Multivariate Fault Prediction (SMFP), for predicting the next multivariate DTC fault in an event sequence, using Long Short-Term Memory Networks (LSTMs) and jointly learned event embeddings. By performing an in-depth comparison of different architectural choices and contextual preprocessing techniques, we provide an initial baseline for SMFP that achieves top-3 accuracy of 63% on predicting multivariate fault with 3 collective output layers, using vehicle maintenance data as a case study.

show abstract

Deep Reinforcement Learning for Auto-optimization of I/O Accelerator Parameters

Pham

Djan

2020

Benchmarking, Measuring, and Optimizing

View full text Add to dashboard Cite

Recurrent Neural Network for Classifying of Hpc Applications

Cited by 5 publications

References 4 publications

The Role of Idle Waves, Desynchronization, and Bottleneck Evasion in the Performance of Parallel Programs

The Role of Idle Waves, Desynchronization, and Bottleneck Evasion in the Performance of Parallel Programs

Towards Sequential Multivariate Fault Prediction for Vehicular Predictive Maintenance

Deep Reinforcement Learning for Auto-optimization of I/O Accelerator Parameters

Contact Info

Product

Resources

About