Cloud Workload Turning Points Prediction via Cloud Feature-Enhanced Deep Learning

Ruan, Li; Li, Shaoning; Lv, Jiaxun; Zhang, Tianyuan; Xiao, Limin; Fang, Haiguang; Wang, Chunhao; Xue, Yunzhi

doi:10.1109/tcc.2022.3160228

Cited by 14 publications

(6 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Correspondingly, the five exclusive workload prediction classes are designated as Evolutionary Learning, Deep Learning, Hybrid Learning, Ensemble Learning, and Quantum Learning. In Evolutionary learning class, the candidate approaches: ANN+SADE [25], ANN-BADE [26], ANN-BHO [27], SDWF [27], and FLGAPSONN [28] [30], Bi-LSTM [31], Crystal ILP [32], and FEMT-LSTM [33] which are pre-dominantly based on functionality of LSTM models. Autoencoder sub-category consists of Encoder+LSTM [34], CP Autoencoder [35], LPAW Autoencoder [36], and GRUED [37] are derived by applying some useful modification in the traditional autoencoders.…”

Section: Methodsmentioning

confidence: 99%

“…The performance is measured as predicted output is achieved for cloud resource management. 2D LSTM [30] Bi-LSTM [31] Crystal ILP [32] FEMT-LSTM [33] Encoder+ LSTM [34] CP Autoen -coder [35] LPAW Autoe -ncoder [36] GRUED [37] DBN+R BN [38] DBN+O ED [39] DP-CU PA [40] es-DNN [41] DNN+ MVM [42] DNN-PPE [43] SG-LSTM [44] ADRL [45] Bi-Hyp -rec [46] BG-LSTM [47] HPF-DNN [48] FAHP [28] ACPS [49] LSRU [50] KSE+WMC [51] FAST [52] SGW-S [19] ClIn [53] AMS [54] E-ELM [55] SF-Cluster [48] EQNN [56] Fig. 3: Classification and Taxonomy of Machine learning based Workload Prediction Models…”

Section: Workload Prediction Operational Flowmentioning

confidence: 99%

“…Unlike traditional LSTM which uses only forward state, Bi-LSTM operates on both forward and backward states to allow more accurate estimation of the weights of both closer and farther input features. Ruan et al [33] have established a turning point prediction model for cloud server workload forecasting considering cloud workload features. Thenafter, a cloud feature-enhanced deep learning model with rulefiltering based Piecewise Linear Representation (PLR) algorithm is build for workload turning point prediction.…”

Section: Lstm-rnnmentioning

confidence: 99%

See 2 more Smart Citations

Performance Analysis of Machine Learning Centered Workload Prediction Models for Cloud

Saxena,

Kumar,

Singh

et al. 2023

Preprint

View full text Add to dashboard Cite

The precise estimation of resource usage is a complex and challenging issue due to the high variability and dimensionality of heterogeneous service types and dynamic workloads. Over the last few years, the prediction of resource usage and traffic has received ample attention from the research community. Many machine learning-based workload forecasting models have been developed by exploiting their computational power and learning capabilities. This paper presents the first systematic survey cum performance analysis-based comparative study of diversified machine learning-driven cloud workload prediction models. The discussion initiates with the significance of predictive resource management followed by a schematic description, operational design, motivation, and challenges concerning these workload prediction models. Classification and taxonomy of different prediction approaches into five distinct categories are presented focusing on the theoretical concepts and mathematical functioning of the existing state-of-the-art workload prediction methods. The most prominent prediction approaches belonging to a distinct class of machine learning models are thoroughly surveyed and compared. All five classified machine learning-based workload prediction models are implemented on a common platform for systematic investigation and comparison using three distinct benchmark cloud workload traces via experimental analysis. The essential key performance indicators of state-of-the-art approaches are evaluated for comparison and the paper is concluded by discussing the trade-offs and notable remarks.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Workload Prediction Operational Flowmentioning

confidence: 99%

Section: Lstm-rnnmentioning

confidence: 99%

See 1 more Smart Citation

Performance Analysis of Machine Learning Centered Workload Prediction Models for Cloud

Saxena,

Kumar,

Singh

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Then it computes a resource allocation for the container u C by using the error and M so that, ideally, τ C = τ • C . Each controller on a machine m transfers to the Supervisor the computed u C that aggregates all the allocations in a vector 3 Ūm , and, if needed, computes a feasible resource allocation u C for each container according to a specified policy (e.g., proportional, priority-based, requirement-based). Moreover, if the sum of to-be-allocated resources is lower than the capacity of the machine, the supervisor can (optionally) scale up the allocations to speed up applications' performance at the expense of a sub-optimal allocation (over-provisioning).…”

Section: Control Architecturementioning

confidence: 99%

“…Software systems are increasingly sophisticated, and they frequently need to handle a wide range of dynamic workloads while maintaining a set service quality [1], [2]. As a result, system scalability is extremely important, and computational resources should be allocated as needed [3], [4], [5]. Provisioned resources should, ideally, match the intensity of dynamic workloads to be served and avoid both underprovisioning (i.e., resources are not enough to handle the workload) and overprovisioning (i.e., resources are more than needed) scenarios [6], [7].…”

Section: Introductionmentioning

confidence: 99%

Autoscaling Solutions for Cloud Applications Under Dynamic Workloads

Quattrocchi,

Incerto,

Pinciroli

et al. 2024

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

Autoscaling systems provide means to automatically change the resources allocated to a software system according to the incoming workload and its actual needs. Public cloud providers offer a variety of autoscaling solutions, ranging from those based on user-written rules to more sophisticated ones. Originally, these solutions were conceived to manage clusters of virtual machines, while more recently, they have also been employed in the operation of containers. This paper analyses the autoscaling solutions provided by three major cloud providers, namely Amazon Web Services, Google Cloud Platform, and Microsoft Azure, and compares them against two solutions we develop based on control theory (ScaleX ) and queuing theory (QN-CTRL). We evaluate the different approaches using both an in-house simulation engine and cloud deployments by feeding them with various synthetic and real-world workloads. Our extensive evaluation collects both simulation results and real measurements by which we can assess that both ScaleX and QN-CTRL outperform industrial techniques in most cases when considering the trade-offs between the service-level-agreement (SLA) violations and the optimal usage of resources.

show abstract

A Deep Learning-Based Algorithm for Predicting the Turning Point of Cloud Workload

Jain,

Panda

2024

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Cloud Workload Turning Points Prediction via Cloud Feature-Enhanced Deep Learning

Cited by 14 publications

References 28 publications

Performance Analysis of Machine Learning Centered Workload Prediction Models for Cloud

Performance Analysis of Machine Learning Centered Workload Prediction Models for Cloud

Autoscaling Solutions for Cloud Applications Under Dynamic Workloads

A Deep Learning-Based Algorithm for Predicting the Turning Point of Cloud Workload

Contact Info

Product

Resources

About