An Efficient Approach for Multi-tenant Elastic Business Processes Management in Cloud Computing Environment

Abstract

Even though the proven benefits of cloud computing paradigm, it must face a serious problem that can compromise its commercial success. It concerns the lack of efficient approach for using optimally the available resources. For this, several approaches have been proposed. However, they suffer from several shortcomings. For instance, often only one objective is taken into account expressing all operations in terms of cost. Furthermore, business processes should be insured with elasticity and multitenancy mechanism while adjusting the available resources to the dynamic load distribution. The proposed approach aims to optimize two conflicting objectives, namely the number of migrated tenants and the cost incurred using a set of resources. It allows to take into account the multi-tenancy property and the Cloud computing elasticity, and is efficient as shown by an extensive experimentation based on real data from Bonita BPM customers.

A Genetic Algorithm for Cost-Aware Business Processes Execution in the Cloud

“…In this paper, we present an evolution of our previous work [2]. We have based our approach on time series segmentation for deducing the "good" time slots to migrate tenants, and on the iterative use of an enhanced version of our time slot heuristic [16]. We also presented the corresponding ILP (Integer Linear Programming) model.…”

Efficient Migration-Aware Algorithms for Elastic BPMaaS

“…We also enhanced our previous timeslot algorithm implementation [2] with the restriction on the tenant list. For performance reasons we don't consider here the subset sum for each number of migrations but only all the tenants.…”

“…Our work is an evolution of our precedent paper [2] where we proposed an bi-objective optimization model for cost and migrations quantity for all tenants scrambled from a timeslot to the next, and a corresponding efficient heuristic. Simply repeating this heuristic on multiple timeslots could provoke the migration of the same tenants, thus provoking breaks of quality of service.…”

“…This algorithm is based on our previous timeslot heuristic [2]. Its principle is to consider that, regarding an initial distribution, we search the best distribution for the next timeslot, knowing that the required capacity of each tenant can change.…”

“…the number of task execution per hour. In our previous work [2], we proposed to optimize resource consumption from a timeslot to the next. In this paper, we extend the method to optimize resource consumption on an arbitrary number of timeslots.…”