Advanced

Model-Based Deadtime Compensation of Virtual Machine Startup Times

Dellkrantz, Manfred LU ; Dürango, Jonas LU ; Robertsson, Anders LU and Kihl, Maria LU (2015) 10th International Workshop on Feedback Computing
Abstract
Scaling the amount of resources allocated to an application according to the actual load is a challenging problem in cloud computing. The emergence of autoscaling techniques allows for autonomous decisions to be taken when to acquire or release resources. The actuation of these decisions is however affected by time delays. Therefore, it becomes critical for the autoscaler to account for this phenomenon, in order to avoid over- or under-provisioning.



This paper presents a delay-compensator inspired by the Smith predictor. The compensator allows one to close a simple feedback loop around a cloud application with a large, time-varying delay, preserving the stability of the controlled system. It also makes it possible for... (More)
Scaling the amount of resources allocated to an application according to the actual load is a challenging problem in cloud computing. The emergence of autoscaling techniques allows for autonomous decisions to be taken when to acquire or release resources. The actuation of these decisions is however affected by time delays. Therefore, it becomes critical for the autoscaler to account for this phenomenon, in order to avoid over- or under-provisioning.



This paper presents a delay-compensator inspired by the Smith predictor. The compensator allows one to close a simple feedback loop around a cloud application with a large, time-varying delay, preserving the stability of the controlled system. It also makes it possible for the closed-loop system to converge to a steady-state, even in presence of resource quantization. The presented approach is compared to a threshold-based controller with a cooldown period, that is typically adopted in industrial applications. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
10th International Workshop on Feedback Computing
language
English
LU publication?
yes
id
96eef5a3-648a-4a30-a77f-f426aec4c51e (old id 5267993)
date added to LUP
2015-04-08 09:09:53
date last changed
2016-06-28 10:54:28
@misc{96eef5a3-648a-4a30-a77f-f426aec4c51e,
  abstract     = {Scaling the amount of resources allocated to an application according to the actual load is a challenging problem in cloud computing. The emergence of autoscaling techniques allows for autonomous decisions to be taken when to acquire or release resources. The actuation of these decisions is however affected by time delays. Therefore, it becomes critical for the autoscaler to account for this phenomenon, in order to avoid over- or under-provisioning.<br/><br>
<br/><br>
This paper presents a delay-compensator inspired by the Smith predictor. The compensator allows one to close a simple feedback loop around a cloud application with a large, time-varying delay, preserving the stability of the controlled system. It also makes it possible for the closed-loop system to converge to a steady-state, even in presence of resource quantization. The presented approach is compared to a threshold-based controller with a cooldown period, that is typically adopted in industrial applications.},
  author       = {Dellkrantz, Manfred and Dürango, Jonas and Robertsson, Anders and Kihl, Maria},
  language     = {eng},
  title        = {Model-Based Deadtime Compensation of Virtual Machine Startup Times},
  year         = {2015},
}