Load Prediction and Hot Spot Detection Models for Autonomic Cloud Computing

Saripalli, Prasad; Kiran, G.V.R.; Shankar, Ravi; Narware, Harish; Bindal, Nitin

doi:10.1109/ucc.2011.66

Cited by 51 publications

(13 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conversely dynamically provisioned infrastructure provided lower hosting costs but with performance caveats resulting from infrastructure launch latency similar to [22]. This key cost versus performance tradeoff for infrastructure provisioning highlights the need for good hot spot detection and load prediction techniques [23].…”

Section: Background and Related Workmentioning

confidence: 99%

Demystifying the Clouds: Harnessing Resource Utilization Models for Cost Effective Infrastructure Alternatives

Lloyd

Pallickara

David

et al. 2017

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

Deployment of Service Oriented Applications (SOAs) to public infrastructure-as-a-service (IaaS) clouds presents challenges to system analysts. Public clouds offer an increasing array of virtual machine types with qualitatively defined CPU, disk, and network I/O capabilities. Determining cost effective application deployments requires selecting both the quantity and type of virtual machine (VM) resources for hosting SOA workloads of interest. Hosting decisions must utilize sufficient infrastructure to meet service level objectives and cope with service demand. To support these decisions, analysts must: (1) understand how their SOA behaves in the cloud; (2) quantify representative workload(s) for execution; and (3) support service level objectives regardless of the performance limits of the hosting infrastructure. In this paper we introduce a workload cost prediction methodology which harnesses operating system time accounting principles to support equivalent SOA workload performance using alternate virtual machine (VM) types. We demonstrate how the use of resource utilization checkpointing supports capturing the total resource utilization profile for SOA workloads executed across a pool of VMs. Given these workload profiles, we develop and evaluate our cost prediction methodology using six SOAs. We demonstrate how our methodology can support finding alternate infrastructures that afford lower hosting costs while offering equal or better performance using any VM type on Amazon's public elastic compute cloud.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Demystifying the Clouds: Harnessing Resource Utilization Models for Cost Effective Infrastructure Alternatives

Lloyd

Pallickara

David

et al. 2017

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

show abstract

“…Some scholars research and propose a kind of forecasting method of cloud computing load resources matched with the mode in this field, such as Caron, et al [2] , and this method used improving the string matching algorithms (KMP). Saripalli, et al [5] proposed a method to identify the resources load peak and forecasting.. Prevost, et al [6] use the neural network and self-regression model to forecast the loads of different types of resources.…”

Section: Related Workmentioning

confidence: 99%

The Energy-Consumption of CPU and Memory Prediction in Cloud Computing Based on GREY-ARIMA Model

2016

Proceedings of 2016 the 6th International Workshop on Computer Science and Engineering

View full text Add to dashboard Cite

The energy-consumption of data center in cloud computing is a heated issues. This paper focuses on the challenges of operation and power management in cloud platform, and presents a GM-ARIMA model, which is simultaneously exploiting autoregressive integrated moving average(ARIMA model) and GM(1,1) model. This combination is used for CPU and Memory in the cloud computing experiment to model, test and forecast. The experiments revealed the GM-ARIMA model, in the minimum amount of history record, is an efficient and feasible model for the limited sampled forecasting.

show abstract

“…In our previous work, [4], we show how slow variations in the Wikipedia workload can be modeled using B-splines in order to extract the seasonality and trend in the workload. In [5], the authors employ an approach where exponential smoothing is used for workload prediction and hot spot detection.…”

Section: Related Workmentioning

confidence: 99%

Online Spike Detection in Cloud Workloads

Mehta

Dürango

Tordsson

et al. 2015

2015 IEEE International Conference on Cloud Engineering

View full text Add to dashboard Cite

Abstract-We investigate methods for detection of rapid workload increases (load spikes) for cloud workloads. Such rapid and unexpected workload spikes are a main cause for poor performance or even crashing applications as the allocated cloud resources become insufficient. To detect the spikes early is fundamental to perform corrective management actions, like allocating additional resources, before the spikes become large enough to cause problems. For this, we propose a number of methods for early spike detection, based on established techniques from adaptive signal processing. A comparative evaluation shows, for example, to what extent the different methods manage to detect the spikes, how early the detection is made, and how frequently they falsely report spikes.

show abstract

Load Prediction and Hot Spot Detection Models for Autonomic Cloud Computing

Cited by 51 publications

References 6 publications

Demystifying the Clouds: Harnessing Resource Utilization Models for Cost Effective Infrastructure Alternatives

Demystifying the Clouds: Harnessing Resource Utilization Models for Cost Effective Infrastructure Alternatives

The Energy-Consumption of CPU and Memory Prediction in Cloud Computing Based on GREY-ARIMA Model

Online Spike Detection in Cloud Workloads

Contact Info

Product

Resources

About