Search

Mobile Grid Services has given the ability to move jobs, data and application software from nodes to nodes during jobs execution in the grid environment. These migrations depend on the grid’s users and the grid’s nodes policies. The heterogeneity, diversity of policies and attributes leads to a need for policy management tools that can handle these problems. Also, before the users can submit their jobs or run their applications on a certain resource or system they may need to guarantee that this resource or system has not been compromised, which could result in their own application or data being stolen or asking for certain users to be allowed to access the service. To date, not enough attention has been paid to policies that deal with such concerns. Most existing grid systems have support only limited types of policies (e.g. CPU resources). A few designs consider enforcing data policies in their architecture. Therefore, we propose a dynamic policy framework that addresses these issues (user-submitted policy, data policy and multiple Virtual Organizations (VOs).

Mobile ad hoc networks (MANETs) are autonomous systems which are comprised of a number of mobile nodes that communicate between themselves by wireless communication in a peer-to-peer basis. They are self-organized, self-configured and self-controlled infrastructure-less networks. Nodes can communicate with each other without any pre-planned or a base station. Disseminating information securely between these nodes in such networks however is a challenging task, particularly when the information is confidential. Revealing such information to anyone else other than the intended nodes could be highly damaging, especially in military applications where keeping the message secret from adversaries is essential. In this paper we present our novel framework for privacy control in mobile ad hoc networks in which privacy policies are attached to messages as they are send between peers. We evaluate our framework using the Network Simulator (NS-2) to provide and check whether the privacy and confidentiality of the originator are met. For this we implemented the privacy enforcement as an NS2 agent that manages and enforces the policies attached to packets at every node in the MANET.

In a multi-organization environment like the GRID, each institute might want to implement limits on how its resources are being used by other institutes. The problem in most traditional policy frameworks lies in the fact that it concentrates on the policy management inside the VO rather than on the management between multiple VOs. The external users, who want to utilize resources in the GRID, ask for guarantees for their jobs and data which most recent policy frameworks do not take into consideration. To date, not enough attention has been paid to policies that deal with such concerns since most existing Grid systems have supported only limited types of policies. This paper introduces a new dynamic framework that supports the management of security policies within the VOs along with the support for the data policies with addressing external users’ policies when making decisions.