Ponder2 - A Policy Environment for Autonomous Pervasive Systems

Abstract

Policies form an important part of management and can be an effective means of implementing self-adaptation in pervasive systems. Most policy-based systems focus on large-scale networks and distributed systems. Consequently, they are often fragmented, dependent on infrastructure and lacking flexibility and extensibility. This paper presents Ponder2, a novel policy system that is suitable for a wide range of environments and applications. The design and implementation of Ponder2 emphasises simplicity, flexibility and extensibility and provides users with the ability to interact easily with the managed system. Ponder2 can interact with other software and hardware components and is being used in environments ranging from single devices, to personal area networks, ad-hoc networks and distributed systems. We also describe PonderTalk, a high-level object orientated language inspired by Smalltalk for configuring and controlling Ponder2 systems.

Low-Latency Service Data Aggregation Using Policy Obligations

“…Informally this means that a policy defines the granularity over time at which data is pushed up the service chain to aggregating services and end-users. In this work the policies are modelled similar to the well-understood Event-Condition-Action paradigm [9], [23]. However, the novelty of the policies used in this work is that they use temporal conditions that describe the distance between two consecutive actions that push data to aggregating services, rather than defining condition on the system state.…”

“…[6] and ECA-based policy languages e.g. [23] are used to control and manage distributed systems [19]. In our work, we are however mainly concerned about the selection and propagation of events.…”

Representation-Independent Data Usage Control

“…In terms of general-purpose usage control models, there are similarities with the models underlying XACML [40], Ponder2 [41] and UCON [42]. The first two, however, do not provide formalized support for cardinality or temporal operators (free text fields exist, but the respective requirements are hard to enforce).…”

Reactive Policies for the Semantic Web

“…Ponder [15] (and its successor Ponder2 [31]) is a policy language for pervasive systems that supports-beyond classical authorization policiesreactive obligation policies. Those obligation policies are specifically used to trigger internal auditing if a security violation happens and are not meant to react to external events.…”