QoS and energy management with Petri nets: A self-adaptive framework

Abstract

Energy use is becoming a key design consideration in computing infrastructures and services. In this paper we focus on service-based applications and we propose an adaptation framework that can be used to reduce power consumption according to the observed workload. The adaptation guarantees a trade-off between energy consumption and system performance. The approach is based on the principle of proportional energy consumption obtained by scaling down energy for unused resources, considering both the number of servers switched on and their operating frequencies. Stochastic Petri Nets are proposed for the modeling of the framework concerns, their analyses give results about the trade-offs. The application of the approach to a simple case study shows its usefulness and practical applicability. Finally, different types of workloads are analyzed with validation purposes.

Designing service-based applications in the presence of non-functional properties: A mapping study

“…Middleware [198], [57], [77], [110], [210], [36], [67], [131], [90], [76], [185] [13], [88], [29], [39], [145], [130], [165], [111], [166], [34], [172], [217], [187], [66], [37], [52], [94], [2], [10], [152], [82], [195], [3], [5], [12], [26], [31], [45], [92], [101], [211], [127], [137], [138], [140], [170], [173], [69], [164].…”

“…Efficiency [196], [43], [8], [167], [34], [217], [42], [60], [189], [86], [179], [1], [51], [204], [93], [140], [178], [175]. Maintainability [1], [51], [93].…”

Software QoS enhancement through self-adaptation and formal models

“…In this tutorial we explain some techniques and research advances for QoS enhancement using formal methods for the adaptation planning (i.e., what to adapt) and adaptation decision (i.e., when to adapt) activities of self-adaptive software [20,4,13,12,14,16,18]. The QoS and non-functional properties that will be illustrated are: service performance, availability, energy consumption of software and running cost.…”

Formal description of an approach for power consumption estimation of embedded systems

“…The resulting model is called Generalized Stochastic Petri Net (GSPN) [16] and is characterized by using only exponentially distributed times. This idea is then developed in [17] and, finally, in [18], where non-functional properties, like energy consumption, are taken into account. The same Petri net type (GSPN) is used in [19] to evaluate the system performance.…”