Efficient distributed monitoring with active Collaborative Prediction

Feng, Dawei; Germain, Cécile; Glatard, Tristan

doi:10.1016/j.future.2013.06.001

Cited by 4 publications

(9 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The goal of this paper is very different: the contribution of SMF and SMFA is on the focus on smarter, data-driven sampling schemes for the case where historical information is available. Rish and Tesauro [2007] proposed a method to handle fault inference based on a collaborative prediction approach, further analyzed in Feng et al [2013]. Although this method significantly reduces the number of required probes for acquiring an accurate view of the system, it is somehow static.…”

Section: Matrix Factorization For Fault Predictionmentioning

confidence: 99%

“…Rish and Tesauro [2007] integrated active learning with MMMF (min-margin heuristic) and [Feng et al 2013] have shown that active learning was a required ingredient in the most difficult and realistic case on a real-world fault prediction example. On the other hand, active learning is somehow inconvenient: because the probe selection is adaptive, it requires a feedback loop, an interface between online analysis and monitoring, and thus a more complicated software than with a predetermined setting.…”

Section: Motivationmentioning

confidence: 99%

“…In active matrix factorization [Rish and Tesauro 2007;Feng et al 2013], the prediction performance is improved by selecting the sample entries in X t actively and iteratively, using the most uncertain heuristic from the very last prediction until the maximum allowed number of samples is reached. The key idea is that, with the progress of each iteration, confidence in the estimation increases simultaneously.…”

Section: Sequential Matrix Factorization With Active Sampling Smfamentioning

confidence: 99%

“…Our previous work [Feng et al 2013], inspired by Rish and Tesauro [2007], addressed fault prediction in a classical collaborative prediction framework, as a purely spatial problem where the matrix is assumed to be a snapshot of the probes outcomes. Here, we take into account the fact that the system dynamically evolves at various time scales.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fault Monitoring with Sequential Matrix Factorization

Feng

Germain

2015

ACM Trans. Auton. Adapt. Syst.

Self Cite

View full text Add to dashboard Cite

For real-world distributed systems, the knowledge component at the core of the MAPE-K loop has to be inferred, as it cannot be realistically assumed to be defined a priori. Accordingly, this paper considers fault monitoring as a latent factors discovery problem. In the context of end-to-end probing, the goal is to devise an efficient sampling policy that makes the best use of a constrained sampling budget.Previous work addresses fault monitoring in a collaborative prediction framework, where the information is a snapshot of the probes outcomes. Here, we take into account the fact that the system dynamically evolves at various time scales. We propose and evaluate Sequential Matrix Factorization (SMF) that exploits both the recent advances in matrix factorization for the instantaneous information and a new sampling heuristics based on historical information. The effectiveness of the SMF approach is exemplified on datasets of increasing difficulty and compared with state of the art history-based or snapshot-based methods. In all cases, strong adaptivity under the specific flavor of active learning is required to unleash the full potential of coupling the most confident and the most uncertain sampling heuristics, which is the cornerstone of SMF.

show abstract

Section: Matrix Factorization For Fault Predictionmentioning

confidence: 99%

Section: Motivationmentioning

confidence: 99%

Section: Sequential Matrix Factorization With Active Sampling Smfamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fault Monitoring with Sequential Matrix Factorization

Feng

Germain

2015

ACM Trans. Auton. Adapt. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our previous work [17], inspired by [18], addressed fault prediction in a classical Collaborative Prediction framework, as a purely spatial problem where the matrix is assumed to be a snapshot of the probes outcomes. Here, we take into account the fact that the system dynamically evolves at various time scales.…”

Section: Introductionmentioning

confidence: 99%

Sequential Fault Monitoring

Feng

Germain

Nauroy

2014

2014 International Conference on Cloud and Autonomic Computing

Self Cite

View full text Add to dashboard Cite

For large-scale distributed systems, the knowledge component at the core of the MAPE-K loop remains elusive. In the context of end-to-end probing, fault monitoring can be recasted as an inference problem in the space-time domain. We propose and evaluate Sequential Matrix Factorization (SMF), a fully spatio-temporal method that exploits both the recent advances in matrix factorization for the spatial information and a new heuristics based on historical information. Adaptivity operates at two levels: algorithmically, as the exploration/exploitation tradeoff is controlled by a self-calibrating parameter; and at the policy level, as active learning is required for the most challenging cases of a real-world dataset.

show abstract

Predicting computational reproducibility of data analysis pipelines in large population studies using collaborative filtering

Barghi

Scaria

Salari

et al. 2018

2018 IEEE International Conference on Big Data (Big Data)

Self Cite

View full text Add to dashboard Cite

Evaluating the computational reproducibility of data analysis pipelines has become a critical issue. It is, however, a cumbersome process for analyses that involve data from large populations of subjects, due to their computational and storage requirements. We present a method to predict the computational reproducibility of data analysis pipelines in large population studies. We formulate the problem as a collaborative filtering process, with constraints on the construction of the training set. We propose 6 different strategies to build the training set, which we evaluate on 2 datasets, a synthetic one modeling a population with a growing number of subject types, and a real one obtained with neuroinformatics pipelines. Results show that one sampling method, "Random File Numbers (Uniform)" is able to predict computational reproducibility with a good accuracy. We also analyze the relevance of including file and subject biases in the collaborative filtering model. We conclude that the proposed method is able to speedup reproducibility evaluations substantially, with a reduced accuracy loss.

show abstract

Efficient distributed monitoring with active Collaborative Prediction

Cited by 4 publications

References 22 publications

Fault Monitoring with Sequential Matrix Factorization

Fault Monitoring with Sequential Matrix Factorization

Sequential Fault Monitoring

Predicting computational reproducibility of data analysis pipelines in large population studies using collaborative filtering

Contact Info

Product

Resources

About