Empirical prediction models for adaptive resource provisioning in the cloud

By Sadeka Islam, Jacky Wai Keung, Kevin Lee, Anna Yun-Wen Liu. Future generations computer systems, 2012

647 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, Information Systems

View Paper: https://www.semanticscholar.org/paper/cd1aa2548a68d1c3b765df0d347e16b72d0615af

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

In Information Systems

Citations in the corpus

A Review of Auto-scaling Techniques for Elastic Applications in Cloud Environments (2014, methodology): “[67] propose using a 12-minute interval, because the setup time of VM instances in the cloud is typically around 5-15 min.”
From Cloud to Fog Computing: A Review and a Conceptual Live VM Migration Framework (2017, methodology): “[103] proposed an adaptive prediction-based resource measurement and provisioning strategy model by combining neural network and linear regression.”
Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms (2017, background): “These works predict resource demand, resource utilization, or job/task length for provisioning or scheduling purposes [4, 9, 12, 14, 23].”
Predicting Cloud Resource Utilization (2016, background): “[16] propose empirical prediction of resource usage in the Cloud.”
Autopilot: workload autoscaling at Google (2020, methodology): “…forecasting methods, such as autoregressive moving average (ARMA) (as in [28] that also uses a job performance model), neural networks (as in [18]), recurrent neural networks as in [9, 20] or a custom forecast as in [13] which demonstrates that Markov-chain based prediction performs better…”
Workload Prediction Using ARIMA Model and Its Impact on Cloud Applications’ QoS (2015, background): “This challenge has been tackled mainly via reactive approaches [3], [4], [5]—which increase or decrease resources when predefined thresholds are reached—or via proactive approaches [6], [7], [8]—which react to future load variations before their occurrence.”
Elasticity in Cloud Computing: State of the Art and Research Challenges (2018, neutral): “For example, [29], [51], [79], [82], [83], [84], [87], [90], [110], [111] use machine learning.”
Data Center Energy Consumption Modeling: A Survey (2016, methodology): “[380] RL ANN, LR Used ANN and LR for resource overbooking Time consuming ANN training process.”