A control theoretical view of cloud elasticity: taxonomy, survey and challenges

Abstract

The lucrative features of cloud computing such as pay-as-you-go pricing model and dynamic resource provisioning (elasticity) attract clients to host their applications over the cloud to save up-front capital expenditure and to reduce the operational cost of the system. However, the efficient management of hired computational resources is a challenging task. Over the last decade, researchers and practitioners made use of various techniques to propose new methods to address cloud elasticity. Amongst many such techniques, control theory emerges as one of the popular methods to implement elasticity. A plethora of research has been undertaken on cloud elasticity including several review papers that summarise various aspects of elasticity. However, the scope of the existing review articles is broad and focused mostly on the highlevel view of the overall research works rather than on the specific details of a particular implementation technique. While considering the importance, suitability and abundance of control theoretical approaches, this paper is a step forward towards a stand-alone review of control theoretic aspects of cloud elasticity. This paper provides a detailed taxonomy comprising of relevant attributes defining the following two perspectives, i.e., control-theory as an implementation technique as well as cloud elasticity as a target application domain. We carry out an exhaustive review of the literature by classifying the existing elasticity solutions using the attributes of control theoretic perspective. The summarized results are further presented by clustering them with respect to the type of control solutions, thus helping in comparison of the related control solutions. In last, a discussion summarizing the pros and cons of each type of control solutions are presented. This discussion is followed by the detail description of various open research challenges in the field.

Robust Dynamic CPU Resource Provisioning in Virtualized Servers

“…Let parameter c denote the desired CPU utilization to CPU allocation ratio, i.e., c = 1/(1 + h), where h is the headroom as used in and defined right after (5). The mRT with respect to parameter c for Kalman filter -SISO, H ∞ -SISO filter and MCC-KF -SISO, is shown in the Fig.…”

“…Towards this end, different autonomic resource management methods have been proposed to dynamically allocate resources across virtualized applications with diverse workload and highly fluctuating workload demands. Autonomic resource management in a virtualized environment using control-based techniques has recently gained significant attention; see [4], [5] and [6] for a survey. One of the most common approaches to control the application performance is by controlling its CPU utilization within the VM; see, e.g., [7] and references therein.…”

Edge Computing Resource Allocation for Dynamic Networks: The DRUID-NET Vision and Perspective

“…With this capacity, System Theory [27] can provide dynamic modeling methodologies, appropriate for Cloud/IoT-based applications. The interesting reader may refer to survey [28] for an extended analysis of control theoretic approaches on performance modeling and cloud elasticity. Close to DRUID-NET concepts, Dechouniotis et al [29] proposed Linear Parameter Varying (LPV) modeling of cloud applications combined with set-theoretic controllers to guarantee a feasible solution of the elasticity in cloud data centers, while Leontiou et al [30] derived fuzzy Takaki-Sugeno models and designed robust controllers to address simultaneously the problems of vertical and horizontal scaling, and load balancing with stability guarantee.…”

Design and evaluation of a biologically-inspired cloud elasticity framework

“…Over the years, researchers and practitioners have proposed many elastic methods using versatile techniques including but not limited to rule-based [4][5][6][7][8], control theory [9][10][11][12][13], fuzzy logic [14,15], optimisation [16][17][18] and machine learning [19,20]. However, despite a large range of existing elasticity research work, the aim of implementing an efficient scaling technique that satisfies the actual demands is still a challenge to achieve [21][22][23].…”