Yathiraj B. Udupi

A Classification-Based Approach to Policy Refinement

Systems are typically designed based on certain high level goals, such as performance and availability. On the other hand, during operation, usually only low level metrics (e.g., CPU utilization) are measured. The system administrators use their expertise to implicitly map bounds on these metrics such that the high level goals are met. The objective of this research is to create an automated and domain independent approach to derive policy bounds on the low level metrics such that the high level goals are met. These policies may be also be used for monitoring the system for goal assessment purposes. The refinement is carried out using a combination of data classification and test-and-development approaches. An ad hoc system is deployed and a dataset containing values of selected metrics is collected by placing appropriate workloads on the system. The policy bounds are derived by applying classification techniques on this dataset. The classification rules are further refined using statistical distributions to arrive at certain low level rules that are useful for system monitoring and to check the system health when it is deployed and running. We show the validity of our approach for an e-commerce auctioning system (RubiS).

Governance of Cross-Organizational Service Agreements: A Policy-Based Approach

Many real-life organizations are hierarchies of largely autonomous, heterogeneous members (individuals or other organizations), often exhibiting rich policies. We restrict our attention to organizations that monitor their environment, collate events, determine compliance of their behaviors with their policies, and potentially act in anticipation of events to ensure the satisfaction of their policies. This paper models cross-organizational service agreements as resulting in the formation of organizations. This paper emphasizes the importance of proactive policy-based governance in organizations (modeled as multiagent systems) and provides a novel architecture supporting policy monitoring, governance, and enactment. This paper provides an initial formalization and discusses the compliance and completeness of behaviors produced from specified policies. To demonstrate the practical utility of this approach, it is implemented using an existing policy engine and messaging middleware.

Multiagent Policy Architecture for Virtual Business Organizations

A virtual organization (VO) is a dynamic collection of entities (individuals, enterprises, and information resources) collaborating on some computational activity. VOs are an emerging means to model, enact, and manage large-scale service computations. VOs consist of autonomous, heterogeneous members, often exhibiting complex behaviors. Thus VOs are a natural match for policy-based approaches. Traditional policy-based frameworks emphasize reactive behaviors, wherein an external request causes a policy engine to compute a response. However, business service settings require richer policies and call for proactive behaviors. A business not only must respond to explicit requests, but also monitor its environment, collate events, and potentially act in anticipation of events in order to ensure that its policies are satisfied. Autonomous, heterogeneous, proactive entities are best modeled as agents and, therefore, VOs are best understood as multiagent systems. Our main contributions are (1) a proactive multiagent policy-based architecture, (2) a hierarchical model of policy monitoring, compliance checking, and enforcement for VOs, and (3) a formalization of VOs. We evaluate our approach using a real business service scenario

Agent-Based Peer-to-Peer Service Networks: A Study of Effectiveness and Structure Evolution

We study peer-to-peer service networks consisting of autonomous agents who seek and provide information services. Agents potentially help each other by giving referrals to guide the search. Each agent autonomously decides whom to contact for a service and whom to provide a service or a referral. Service networks evolve as the agents change their neighbors to improve how their needs are fulfilled. If an agent autonomously decides to, it may cache some responses from other (information) services. We observe that even a small cache improves agentsý success in discovering needed services and enables a few initial service providers to serve the information needs of many. Caching induces clustering of agents based on interest.

Trustworthy Service Caching: Cooperative Search in P2P Information Systems

We are developing an approach for P2P information systems, where the peers are modelled as autonomous agents. Agents provide services or give referrals to one another to help find trustworthy services. We consider the important case of information services that can be cached. Agents request information services through high-level queries, not by describing specific objects as in caching in traditional distributed systems. Moreover, the agents autonomously decide whom to contact for a service, whom to provide a service or referral, whether to follow a referral and whether to cache a service. Thus the information system itself evolves as agents learn about each other and the contents of the caches of the agents change. We study here the effect of caching on service location and on the information system itself. Our main results are that, (1) even with a small cache, agents can locate services more easily; (2) since the agents that cache services can act like service providers, a small number of initial service providers are enough to serve the information needs of the consumers; and (3) agents benefit from being neighbours with others who have similar interests.

Design Patterns for Policy-Based Service Engagements

Service engagements arise commonly in business and scientific computing. A service engagement is characterized by autonomous parties coming together in a contractual arrangement to share resources or carry out tasks for one another. Recent work shows how to model service engagements in an interactive manner and at a high level. This work formalizes the atoms of a service engagement as commitments among the participants, to be created and manipulated as the engagement progresses. Further, it scopes the commitments of an engagement in a (virtual) organization, and specifies how the policies of the participants affect their interactions. This paper contributes design patterns for service engagements formulated in terms of roles, commitments, and allied concepts. Each pattern reflects a distinct element of a service engagement from a business perspective and highlights exactly where policies apply. This enables the perspicuous, reusable specification of service engagements.

InterPol: a policy framework for managing trust and privacy in referral networks

Referral networks are a kind of P2P system consisting of autonomous agents who seek, provide services, or refer other service providers. Key applications include service discovery and selection, and knowledge sharing. This use of referrals is inspired by human interactions, where referrals are a key basis for judging the trustworthiness of a given service. The use of referrals enable an agent to control how its request is processed, it also provides an architectural basis for four kinds of interaction policies. InterPol is a language and framework supporting such policies. InterPol has been implemented using a Datalog-based policy engine for each agent. It has been applied on scenarios from a (multinational) health care project. The contribution of this paper is in a general referrals-based framework for privacy and trust management, which is shown to effectively capture a variety of privacy and trust requirements of autonomous users.