A Comprehensive Context for Mobile-Code Deployment
Abstract
Given the acknowledged importance of mobile code, current distribution models are surprisingly primitive. For example, Java's model assumes that the constituent parts that make up a mobile program will all be downloaded to a single location, and then verified, linked, possibly dynamically compiled, and finally executed at that very location. This research project made three important contributions: First, it demonstrated that it can be beneficial to perform verification, dynamic compilation, and execution at different physical locations. A prototype was built that performs code verification and just-in-time compilation at a "code generating router" inside the network itself. If the end-points of the network are resource-limited devices such as wirelessly-connected personal digital assistants (PDA's), off-loading dynamic code generation to the stationary network can result in substantial benefits. As a second contribution, the project identified a novel attack on mobile code systems, based on the complexity of the code verification algorithm itself. The third contribution is a new mobile-code verification algorithm that not only lacks this vulnerability, but that is also more efficient.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2005
- Accession Number
- ADA437278
Entities
People
- Michael Franz
Organizations
- University of California, Irvine