Network Device Workload Prediction: A Data Mining Challenge at Knowledge Pit

Janusz, Andrzej; Przyborowski, Mateusz; Biczyk, Piotr; Ślęzak, Dominik

doi:10.15439/2020f159

Cited by 16 publications

(20 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A slightly different approach is to use Japanese candles as summaries [9], and then compute and process that data. One example is the annual AI competition, which this year is based on such summaries [6].…”

Section: B Goal Descriptionmentioning

confidence: 99%

Generating Fuzzy Linguistic Summaries for Menstrual Cycles

Sosnowski

Penza²

2020

Proceedings of the 2020 Federated Conference on Computer Science and Information Systems

View full text Add to dashboard Cite

This paper presents a method of generating linguistic summaries of women's menstrual cycles based on the set of concepts describing various aspects of the cycles. These concepts enable description of menstrual cycles that are readable for humans, but they also provide high-level information that can be used as control input for other data processing actions such as e.g. anomaly detection. The labels signifying these concepts are assigned to cycles by means of multivariate time series analysis. The corresponding algorithm is a subsystem of a bigger solution created as a part of an R&D project.

show abstract

Section: B Goal Descriptionmentioning

confidence: 99%

Generating Fuzzy Linguistic Summaries for Menstrual Cycles

Sosnowski

Penza²

2020

Proceedings of the 2020 Federated Conference on Computer Science and Information Systems

View full text Add to dashboard Cite

show abstract

“…The reconstruction of such long data series is typically highly error-prone and would therefore worsen our model. In addition, missing data at the beginning is not critical, since it merely shortens the 1 GitHub link to Telescope: https://github.com/DescartesResearch/telescope time series. To impute the missing values within the time series, we assume a daily pattern within the data.…”

Section: A Missing Data Imputationmentioning

confidence: 99%

“…To overcome this problem, proactive adaptation algorithms are required that analyze historical data and automatically forecast future conditions to enable early decision making. However, the decision making component is beyond the scope of this paper, as this paper is part of the FedCSIS 2020 Network Device Workload Prediction Challenge [1]. To achieve sufficient forecasting performance, no single method can be used since the "No-Free-Lunch-Theorem" states that there cannot be a single algorithm that outperforms all others on every kind of data [2].…”

Section: Introductionmentioning

confidence: 99%

A Framework for Time Series Preprocessing and History-based Forecasting Method Recommendation

Züfle

Kounev

2020

Proceedings of the 2020 Federated Conference on Computer Science and Information Systems

View full text Add to dashboard Cite

The complexity of managing the capacities of large IT infrastructures is constantly increasing as more network devices are connected. This task can no longer be performed manually, so the system must be monitored at runtime and estimations of future conditions must be made automatically. However, since using a single forecasting method typically performs poorly, this paper presents a framework for forecasting univariate network device workload traces using multiple forecasting methods. First, the time series are preprocessed by imputing missing data and removing anomalies. Then, different features are derived from the univariate time series, depending on the type of forecasting method. In addition, a recommendation approach for selecting the most suitable forecasting method from this set of algorithms for each time series based only on its historical values is proposed. For this purpose, the performance of the forecasting methods is approximated using the historical data of the respective time series under consideration. The framework is used in the FedCSIS 2020 Challenge and shows good forecasting quality with an average R 2 score of 0.2575 on the small test data set.

show abstract

“…Considering this, we present our solution for this years FedCSIS 2020 challenge [14], which is a model for network device workload prediction. It combines the overall average of each KPI series with a prediction from a linear neural network.…”

Section: Introductionmentioning

confidence: 99%

“…Finally section V concludes our paper. [14] was to predict the future workload of network devices based on past workload observations. More specifically, the workload of a set of devices, referred to as hosts, were characterized by KPI series such as CPU utilization, incoming and outgoing network traffic or allocated main memory.…”

Section: Introductionmentioning

confidence: 99%

Superiority of Simplicity: A Lightweight Model for Network Device Workload Prediction

Acker

Wittkopp

Nedelkoski

et al. 2020

Proceedings of the 2020 Federated Conference on Computer Science and Information Systems

View full text Add to dashboard Cite

The rapid growth and distribution of IT systems increases their complexity and aggravates operation and maintenance. To sustain control over large sets of hosts and the connecting networks, monitoring solutions are employed and constantly enhanced. They collect diverse key performance indicators (KPIs) (e.g. CPU utilization, allocated memory, etc.) and provide detailed information about the system state. Predicting the future progress of those KPIs allows ahead of time optimizations like anomaly detection or predictive maintenance and can be defined as a time series forecasting problem. Although, a variety of time series forecasting methods exist, forecasting the progress of IT system KPIs is very hard. First, KPI types like CPU utilization or allocated memory are very different and hard to be modelled by the same model. Second, system components are interconnected and constantly changing due to soft-or firmware updates and hardware modernization. Thus a frequent model retraining or fine-tuning must be expected. Therefore, we propose a lightweight solution for KPI series forecasting. It consists of a weighted heterogeneous ensemble method composed of two models -a neural network and a mean predictor. As ensemble method a weighted summation is used, whereby a heuristic is employed to set the weights. The modelling approach is evaluated on the available FedCSIS 2020 challenge dataset and achieves an overall R 2 score of 0.10 on the preliminary 10% test data and 0.15 on the complete test data. We publish our code on the following github repository: https://github.com/citlab/fed_challenge 10 PROCEEDINGS OF THE FEDCSIS. SOFIA, 2020

show abstract

Network Device Workload Prediction: A Data Mining Challenge at Knowledge Pit

Cited by 16 publications

References 10 publications

Generating Fuzzy Linguistic Summaries for Menstrual Cycles

Generating Fuzzy Linguistic Summaries for Menstrual Cycles

A Framework for Time Series Preprocessing and History-based Forecasting Method Recommendation

Superiority of Simplicity: A Lightweight Model for Network Device Workload Prediction

Contact Info

Product

Resources

About