Loading...

Modeling and evaluation of dispatching policies in IaaS cloud data centers using SANs

Ataie, E ; Sharif University of Technology | 2022

22 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.suscom.2021.100617
  3. Publisher: Elsevier Inc , 2022
  4. Abstract:
  5. Infrastructure-as-a-Service (IaaS) clouds not only have to meet business requirements but also need to consider other important metrics that influence the quality of service, such as performance, availability, and power consumption. In this paper, we present a Stochastic Activity Network (SAN) based analytical approach that simultaneously computes these metrics under a given load for IaaS cloud data centers. The model's high-level abstraction of IaaS clouds allows us to evaluate different resource management strategies and features, including cloud federation and dispatching policies. In a first step, we model a single rack including its autonomous local manager responsible for scheduling IaaS requests onto available PMs. In a second step, we present a unified model that represents an entire data center, including an energy-aware central manager, to take advantage of structure-awareness for managing cloud resources in an optimized way. We introduce and evaluate several dispatching policies that can be used by the cloud central manager to demonstrate the applicability of the unified SAN model. The model is validated against the well-known CloudSim framework. Extended simulations are also conducted to apply the proposed model to a real-world IaaS cloud. © 2021 Elsevier Inc
  6. Keywords:
  7. Analytical modeling ; Dispatching policies ; IaaS cloud ; Performance evaluation ; Automata theory ; Infrastructure as a service (IaaS) ; Managers ; Power management ; Quality of service ; Activity network ; Business requirement ; Cloud data centers ; Dispatching policy ; Infrastructure-as-a-service cloud ; Performances evaluation ; Quality-of-service ; Service clouds ; Stochastic activity network ; Stochastics ; Stochastic systems
  8. Source: Sustainable Computing: Informatics and Systems ; Volume 33 , 2022 ; 22105379 (ISSN)
  9. URL: https://www.sciencedirect.com/science/article/abs/pii/S2210537921001050