Formal Verification of Service Level Agreements Through Distributed Monitoring

Abstract

Abstract. In this paper, we introduce a formal model of the availability, budget compliance and sustainability of distributed services, where service sustainability is a new concept which arises as the composition of service availability and budget compliance. The model formalizes a distributed platform for monitoring the above service characteristics in terms of a parallel composition of task automata, where dynamically generated tasks model asynchronous events with deadlines. The main result of this paper is a formal model to optimize and reason about service characteristics through monitoring. In particular, we use schedulability analysis of the underlying timed automata to optimize and guarantee service sustainability.

Self-Reconfiguring Microservices

“…On a different level, we plan to integrate in our system an existing monitoring functionality of Jolie Enterprise in order to be able to determine the current load of the system and therefore to be able to automatically balance the load, possibly modifying the configuration, in order to maintain some given service level agreements for the deployed services. Suitable extensions of such a monitoring tool could also be used to combine run time checking with static analysis (e.g., based on types) in order to ensure the correctness of the system, and more generally to verify service level agreements along the lines described in [9,37]. The same techniques can be also exploited to automatize the deployment of system developed by means of choreographic languages [17,41,42].…”

Statically and Dynamically Verifiable SLA Metrics

“…There have been several attempts to formalize SLAs, using techniques ranging from semantic annotations [17], to rewriting logics [19] and to constraint programming [5]. In this paper, following [18], we use a very simple formalization based on service metric functions. Service metric functions aggregate a set of basic measurements into a single number that indicates the quality of a certain service characteristic.…”

“…Subsequently, if an SLA mismatch is observed, in the reaction phase, the number of allocated resources are increased. The monitoring platform developed in the Envisage Project also allows to decrease the number of resources if it is too costly/high [18]. The following reaction method verifies every 300s whether the percentage of timely service provisioning requests is reached and, in case of failures, adds one more CPU: Correctness of the monitoring framework (i.e.…”

“…The following reaction method verifies every 300s whether the percentage of timely service provisioning requests is reached and, in case of failures, adds one more CPU: Correctness of the monitoring framework (i.e. that the monitors converge within a user-given time towards the service level objectives specified in an SLA) was investigated in [18]. The idea is to translate the code for the program including the monitoring code into timed automata for use with UPPAAL [4].…”