Towards Inter-cloud Schedulers: A Survey of Meta-scheduling Approaches
Stelios Sotiriadis, Nik Bessis, Nikolaos Antonopoulos · 2011
Cloud computing has emerged as one of the latest technologies for delivering on-demand advanced services over the internet. Various cloud providers have developed data-centres which are spread at several geographically locations, and are available for utilization from internet users. However, as the number of resource consumers is increasing significantly, it becomes apparent that the capacity-oriented clouds require coming together and agreeing on common acting behaviours for improving the quality of service, hence providing an overall optimal load allocation. In this direction, current solutions do not support a coordinated distribution of different cloud workloads. Even geographically distributed data-centres from the same vendor (e.g. Amazon) don't support a seamless mechanic for balancing hosted services as the users require indicating their selected hosts' location. To answer this limitation, a recently emerged inter-clouds notion comes to expand cloud capabilities and to offer an improved sharing paradigm of workloads. Herein we present a state-of-the-art review with a particular focus on the adoptability of current meta-schedulers for managing workloads, towards the inter-cloud era. Specifically, by exploiting inter-cloud requirements (e.g. flexibility, geographically distribution etc.) we evaluate various meta-schedulers for future inter-clouds.