Data augmentation through multivariate scenario forecasting in Data Centers using Generative Adversarial Networks (2022-01-12T00:00:00.000000Z)

TL;DR

This work combines the disciplines of GAN-based data augmentation and scenario forecasting, filling the gap in the generation of synthetic data in DCs, and proposes a methodology to increase the variability and heterogeneity of the generated data by introducing on-demand anomalies without additional effort or expert knowledge.

Abstract

The Cloud paradigm is at a critical point in which the existing energy-efficiency techniques are reaching a plateau, while the computing resources demand at Data Center facilities continues to increase exponentially. The main challenge in achieving a global energy efficiency strategy based on Artificial Intelligence is that we need massive amounts of data to feed the algorithms. This paper proposes a time-series data augmentation methodology based on synthetic scenario forecasting within the Data Center. For this purpose, we will implement a powerful generative algorithm: Generative Adversarial Networks (GANs). Specifically, our work combines the disciplines of GAN-based data augmentation and scenario forecasting, filling the gap in the generation of synthetic data in DCs. Furthermore, we propose a methodology to increase the variability and heterogeneity of the generated data by introducing on-demand anomalies without additional effort or expert knowledge. We also suggest the use of Kullback-Leibler Divergence and Mean Squared Error as new metrics in the validation of synthetic time series generation, as they provide a better overall comparison of multivariate data distributions. We validate our approach using real data collected in an operating Data Center, successfully generating synthetic data helpful for prediction and optimization models. Our research will help optimize the energy consumed in Data Centers, although the proposed methodology can be employed in any similar time-series-like problem.

Authors

Jaime P'erez

1 papers

Patricia Arroba

1 papers

Jose M. Moya

1 papers

TL;DR

Abstract

Authors

References93 items

Generative Adversarial Networks

A deep multivariate time series multistep forecasting network

TSA-GAN: A Robust Generative Adversarial Networks for Time Series Augmentation

Fault Diagnosis of Hydraulic Systems Based on Deep Learning Model With Multirate Data Samples

AdaBelief Optimizer: Adapting Stepsizes by the Belief in Observed Gradients

Renewable scenario generation using stable and controllable generative adversarial networks with transparent latent space

Improving the Accuracy of Global Forecasting Models using Time Series Data Augmentation

An empirical survey of data augmentation for time series classification with neural networks

Conditional Sig-Wasserstein GANs for Time Series Generation

Training Generative Adversarial Networks with Limited Data

Simulation-based assessment of data center waste heat utilization using aquifer thermal energy storage of a university campus

Synthetic Sensor Data for Human Activity Recognition

Recalibrating global data center energy-use estimates

Generating Energy Data for Machine Learning with Recurrent Generative Adversarial Networks

Bayesian forecasting of multivariate time series: scalability, structure uncertainty and decisions

Predictive GPU-based ADAS Management in Energy-Conscious Smart Cities

Using GANs for Sharing Networked Time Series Data: Challenges, Initial Promise, and Open Questions

A survey on Image Data Augmentation for Deep Learning

Biosignal Generation and Latent Variable Analysis With Recurrent Generative Adversarial Networks

A Comparison of ARIMA and LSTM in Forecasting Time Series

Thermal Prediction for Immersion Cooling Data Centers Based on Recurrent Neural Networks

T-CGAN: Conditional Generative Adversarial Network for Data Augmentation in Noisy Time Series with Irregular Sampling

Generative Adversarial Network for Synthetic Time Series Data Generation in Smart Grids

Demand Side Data Generating Based on Conditional Generative Adversarial Networks

How to stop data centres from gobbling up the world’s electricity

Synthetic Sensor Data Generation for Health Applications: A Supervised Deep Learning Approach

Relational recurrent neural networks

Improved Mixed-Example Data Augmentation

A Complete Model for Modular Simulation of Data Centre Power Load

Assessing ICT global emissions footprint: Trends to 2040 & recommendations

SCORE: Simulator for cloud optimization of resources and energy consumption

Spectral Normalization for Generative Adversarial Networks

The Effectiveness of Data Augmentation in Image Classification using Deep Learning

An Unsupervised Deep Learning Approach for Scenario Forecasts

Trends in Data Centre Energy Consumption under the European Code of Conduct for Data Centre Energy Efficiency

Unsupervised Representation Learning with Deep Convolutional Neural Network for Remote Sensing Images

Wasserstein Generative Adversarial Networks

GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs

Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks

Stock market's price movement prediction with LSTM neural networks

Scenario Reduction With Submodular Optimization

Improved Training of Wasserstein GANs

TS-LSTM and Temporal-Inception: Exploiting Spatiotemporal Dynamics for Activity Recognition

English Conversational Telephone Speech Recognition by Humans and Machines

SenseGen: A deep learning architecture for synthetic sensor data generation

Multivariate Industrial Time Series with Cyber-Attack Simulation: Fault Detection Using an LSTM-based Predictive Data Model

Data Augmentation for Time Series Classification using Convolutional Neural Networks

Deep Residual Learning for Image Recognition

GloudSim: Google trace based cloud simulator with virtual machines

Workload modeling for resource usage analysis and simulation in cloud computing

Simulation of the Thermosyphon Free Cooling Mode in an Integrated System of Mechanical Refrigeration and Thermosyphon for Data Centers

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Adam: A Method for Stochastic Optimization

Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank

Predicting the Present with Bayesian Structural Time Series

Comparison of the ARMA, ARIMA, and the autoregressive artificial neural network models in forecasting the monthly inflow of Dez dam reservoir

Stock market prediction using Hidden Markov Models

Bayesian Time Series Models

Bayesian Multivariate Time Series Methods for Empirical Macroeconomics

Thermophysical and transport properties of humid air at temperature range between 0 and 100 °C

Stock market forecasting using hidden Markov model: a new approach

Day-ahead electricity price forecasting using the wavelet transform and ARIMA models

Modeling and forecasting electricity prices with input/output hidden Markov models

ARIMA Models to Predict Next-Day Electricity Prices

Applied Bayesian Forecasting and Time Series Analysis

Generating Diverse Synthetic Medical Image Data for Training Machine Learning Models, 2020

Typical wind power scenario generation for multiple wind farms using conditional improved Wasserstein generative adversarial network

Modernization of Statistical Disclosure Limitation at US Census Bureau

Time-series Generative Adversarial Networks

A simulation study on heat recovery of data center: A case study in Harbin, China

Window-Warping: A Time Series Data Augmentation of IMU Data for Construction Equipment Activity Identification

Global Internet Phenomena

Synthetic Data will Drive Next Wave of Business Applications | GTC Silicon Valley

Scenario Generation for Wind Power Using Improved Generative Adversarial Networks

Generative adversarial network for synthetic time series data 1485 Data augmentation through multivariate scenario forecasting in Data Centers using Generative generation in smart grids