FCMS: A fuzzy controller for CPU and memory consolidation under SLA constraints

Abstract

Summary Cloud providers (CPs) rely on server consolidation (the allocation of several virtual machines [VMs] on the same physical server) to minimize their costs. Maximizing the consolidation level is thus become 1 of the major goals of cloud providers. This is a challenging task because it requires the ability of estimating, in a resource contention scenario, multidimensional resource demands for multitier cloud applications that must meet service‐level agreements (SLAs) in face of nonstationary workloads. In this paper, we cope with the problem of jointly allocating CPU and memory capacity to (a) precisely estimate their capacity required by each VM to meet its SLAs and (b) coordinate their allocation to limit the negative effects due to the interactions of dynamic allocation mechanisms, which, if ignored, can lead to SLA violations. We tackle this problem by devising FCMS, a feedback fuzzy controller that is able to dynamically adjust the CPU and memory capacity allocated to each VM in a coordinated way, to precisely match the needs induced by the incoming workload. By means of an extensive experimental evaluation, we show that FCMS is able to achieve the above goals and works better than existing state‐of‐the‐art alternative solution in all the considered experimental scenarios.

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

“…Furthermore, they do not consider system resource utilisation level. [30,[68][69][70][71]. Apart from Response time and Throughput, the use of some other performance metrics is also observed.…”

“…However, in terms of vertical elasticity, the existing literature lacks on providing the justification of the choice of Control input. It is evident from Tables 1, 2 [32, 34, 55-57, 59, 63-66, 68, 70, 73, 75, 76] and as combined with Memory allocation in [30,42,45,71], combined with bandwidth in [62,77] and combined with both memory and bandwidth in [30,69]. In last, the control solutions proposed in [31,53,54] …”

“…In contrast to the approaches mentioned above, the control solutions proposed in [70,71,76,77] focused on vertical elasticity. Xu et al [76] used two fuzzy based methods, i.e., modelling and prediction to dynamically estimate the CPU capacity of a VM needed by an application.…”

“…[70] focused on multi-tier application to dynamically allocate CPU share of VMs of each tier. This approach is further extended in [71] to also take into account the memory readjustment in collaboration to CPU capacity. Moreover, they also provided a cascade based coordination mechanism at the time of scaling decision to consider the joint effect of both kinds of resources.…”

Prometheus: A flexible toolkit for the experimentation with virtualized infrastructures

“…In order to illustrate how Prometheus can be used in practice, in this section, we discuss how to configure and deploy a testbed by using as a case study the instantiation of Prometheus that we exploit in the work of Anglano et al to experimentally evaluate a resource management strategy named FCMS.…”