A framework for dependable QoS adaptation in probabilistic environments

Abstract

Distributed protocols executing in uncertain environments, like the Internet, had better adapt dynamically to environment changes in order to preserve QoS. In a previous work, it was shown that QoS adaptation should be dependable, if correctness of protocol properties is to be maintained. In this paper we provide concrete strategies and methodologies to improve the implementation of dependable QoS adaptation. During its lifetime, a system alternates periods where its temporal behavior is well characterized, with transition periods where a variation of the environment conditions occurs. Our method is based on the following: if the environment is generically characterized in analytical terms, and we can detect the alternation of these stable and transient phases, we can drastically improve the effectiveness of dependable QoS adaptation. To prove our point, we conduct an evaluation based on "synthetic" data flows generated from one or more probabilistic distributions, and we show that the proposed strategies can indeed be effective and still dependable in the considered cases.

Providing adaptive QoS for Real-Time Publish-Subscribe Service in SGIOC-HQ

“…Casimiro et al [7] provides concrete strategies and methodologies to improve the implementation of dependable QoS adaptation. Their method can drastically improve the effectiveness of dependable QoS adaptation by using alternation of these stable and transient phases.…”

“…Since the environment is probabilistic but the distribution is unknown, we now make the following simple sufficient stability assumption [7]: probabilistic changes are slower than the detection and recognition speed. In other words, we assume that typical execution environments are sufficiently stable for the time it takes to recognize the actual probabilistic distribution.…”

“…For the model that describes the data is known, the problem is simplified to estimating unknown parameters of a known model from the available data. In our experimental implementation of the frame-work, we used the method of maximum likelihood estimation (MLE), which is considered to be one of the most robust techniques for parameter estimation [7,11].Once the most fitting distribution (together with its parameters) has been identified, its statistics properties can be exploited to find proper QoS policy setting that will satisfy the objective of keeping a constant QoS level.…”

Design and development of a proof-of-concept platooning application using the HIDENETS architecture

“…Using this service, any changes in the communication environment affecting communication delays will be reflected in the considered value and hence in the algorithm deadlines. A detailed description of the probabilistic framework used in this service can be found in [2].…”

Adaptare

“…In earlier work [Casimiro et al 2008], we firstly considered that during its lifetime a system alternates periods where its stochastic behavior is well characterized, with transition periods where a variation of the environment conditions occurs. The initial experiments and results were motivating and led to the development of the Adaptare framework.…”