A Debt-Aware Learning Approach for Resource Adaptations in Cloud Elasticity Management

Abstract

Elasticity is a cloud property that enables applications and its execution systems to dynamically acquire and release shared computational resources on demand. Moreover, it unfolds the advantage of economies of scale in the cloud through a drop in the average costs of these shared resources. However, it is still an open challenge to achieve a perfect match between resource demand and provision in autonomous elasticity management. Resource adaptation decisions essentially involve a trade-off between economics and performance, which produces a gap between the ideal and actual resource provisioning. This gap, if not properly managed, can negatively impact the aggregate utility of a cloud customer in the long run. To address this limitation, we propose a technical debt-aware learning approach for autonomous elasticity management based on a reinforcement learning of elasticity debts in resource provisioning; the adaptation pursues strategic decisions that trades off economics against performance. We extend CloudSim and Burlap to evaluate our approach. The evaluation shows that a reinforcement learning of technical debts in elasticity obtains a higher utility for a cloud customer, while conforming expected levels of performance.

Model-Driven Simulation of Elastic OCCI Cloud Resources

“…In the same context, PERFSCALE [32] simulation tool allows simulating the behavior of an elastic application Moreover, a few works have been done [33,18,34] on extending CloudSim for elasticity.…”

“…In [33] the authors propose a technical debt-aware learning approach for autonomous elasticity management based on a reinforcement learning of elasticity debts in resource provisioning. The authors evaluate their approach by extending CloudSim.…”

“…The proposed MDE approach for cloud simulation based on OCCIware metamodel in this article is an initial step in this direction. [33,18,34] No No Yes No CloudMF [36] Yes No No No CMCloudSim [37] No No No Yes SimGrid [38], B. Javadi et al [39] No No No Limited EMDCloudSim [18] No No Yes Limited OtherSim [40,41,42,43,44] No Yes No No In this article, we have proposed a model-driven engineering approach based on the OCCIware metamodel and CloudSim toolkit. This approach takes the advantage of the MDE approach by allowing the simulation of any kind of cloud computing resources and by enabling the users to extend the simulation tool for other capabilities in a high level of abstraction and to model, edit and generate their application with minimum effort and time consuming.…”

Agentes Inteligentes baseados em Aprendizado por Reforço para Alocação Dinâmica de Tráfego em Nuvens

“…Mera-Gomez et al propõem uma abordagem baseada em Q-learning, semelhantè a de Russel e Zindars. Os estados são modelados de acordo com a proporção de clientes com uma ou mais requisições na fila e a priorizaçãoé realizada de acordo com a distância do período de tarifação [Mera-Gómez et al, 2017]. Diferentemente, neste artigo os estados são baseados inteiramente na banda contratada.…”