User behaviour trust model to defend denial of service attacks in distributed computational environments

Today, internet with its innumerable resources has become a basic amenity to the sophisticated society. Major accessibility paradigms used in business and informational domains are web and grid services. Any service is viewed as interoperable interactions among the hosts over a network. Many web and grid resources are vulnerable to security attacks and distributed denial of service attack is a threatening one that makes the resources unavailable. This paper addresses this security issue with respect to the trustworthiness of the client nodes participating in internet computing. Proposed solution is focused at making the underlying network reliable for transactions by identifying attack clients based on their behaviour in request participation using a trust model, to substantially prevent them from participating in normal data transactions and to ensure sustainable network. The model is tested with subnets of varying size and varying amount of request traffic and is deployed in active networking environment.