Cloud Workload Prediction by Means of Simulations

Kecskeméti, Gábor; Kertész, Attila; Németh, Zsolt

doi:10.1145/3075564.3075589

Cited by 4 publications

(2 citation statements)

References 7 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results represented that the model accuracy is up to 91%, with efficiency in resource usage and maintaining the QoS. Kecskemeti et al suggested a workload prediction method using simulations, which provides the knowledge of the underlying clouds to support activities such as cloud orchestration or workflow enactment.…”

Section: Related Workmentioning

confidence: 99%

An autonomic risk‐ and penalty‐aware resource allocation with probabilistic resource scaling mechanism for multilayer cloud resource provisioning

Mazidi

Golsorkhtabaramiri

Tabari

2020

Int J Communication

View full text Add to dashboard Cite

Cloud computing environment allows presenting different services on the Internet in exchange for cost payment. Cloud providers can minimize their operational costs by auto-scaling of the computational resources based on demand received from users. However, the time and cost required to increase and decrease the number of active computational resources are among the biggest limitations of scalability. Thus, auto-scaling is considered as one of the most important challenges in the field of cloud computing.The present study aimed to present a new solution to automatic scalability of resources for multilayered cloud applications under the Monitor-Analysis-Plan-Execute-Knowledge loop. In addition, the Google penalty payment model was used to model the penalty costs in the problem and to accurately evaluate the earned profit. A hybrid resource load prediction algorithm was proposed to evaluate the future of resources in each cloud layer. Further, we used statistical solution to determine the statuses of VMs in addition to presenting a risk-aware algorithm to allocate the user requests to active resources. The experimental results by Cloudsim indicated the improvement of the proposed approach in terms of operational costs, the number of used resources, and the amount of profit. K E Y W O R D Scloud computing, multilayered application, operational cost, scalability

show abstract

Section: Related Workmentioning

confidence: 99%

An autonomic risk‐ and penalty‐aware resource allocation with probabilistic resource scaling mechanism for multilayer cloud resource provisioning

Mazidi

Golsorkhtabaramiri

Tabari

2020

Int J Communication

View full text Add to dashboard Cite

show abstract

“…Many studies have been conducted on various predictions in cloud computing. From the perspective of research objectives, some researchers have studied server load prediction [ [6][7][8][9][10], VM load prediction [11,12], VM utilization prediction [13,14], host utilization prediction [15], web application workload prediction [16], cloud service workload prediction [17][18][19], workflow workload prediction [20], service quality prediction [21], and workload characterization [22][23][24]. Toumi et al [6] described a server load according to the submitted task types and the submission rate and applied a stream mining technique to predict server loads.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Method for Short-Term Host Utilization Prediction in Cloud Computing

Jing

Wang

2019

Journal of Electrical and Computer Engineering

View full text Add to dashboard Cite

Dynamic resource scheduling is a critical activity to guarantee quality of service (QoS) in cloud computing. One challenging problem is how to predict future host utilization in real time. By predicting future host utilization, a cloud data center can place virtual machines to suitable hosts or migrate virtual machines in advance from overloaded or underloaded hosts to guarantee QoS or save energy. However, it is very difficult to accurately predict host utilization in a timely manner because host utilization varies very quickly and exhibits strong instability with many bursts. Although machine learning methods can accurately predict host utilization, it usually takes too much time to ensure rapid resource allocation and scheduling. In this paper, we propose a hybrid method, EEMD-RT-ARIMA, for short-term host utilization prediction based on ensemble empirical mode decomposition (EEMD), runs test (RT), and autoregressive integrated moving average (ARIMA). First, the EEMD method is used to decompose the nonstationary host utilization sequence into relatively stable intrinsic mode function (IMF) components and a residual component to improve prediction accuracy. Then, efficient IMF components are selected and then reconstructed into three new components to reduce the prediction time and error accumulation due to too many IMF components. Finally, the overall prediction results are obtained by superposing the prediction results of three new components, each of which is predicted by the ARIMA method. An experiment is conducted on real host utilization traces from a cloud platform. We compare our method with the ARIMA model and the EEMD-ARIMA method in terms of error, effectiveness, and time-cost analysis. The results show that our method is a cost-effective method and is more suitable for short-term host utilization prediction in cloud computing.

show abstract