Empirical Survey: Experimentation and Implementations of High Speed Protocol Data Transfer for GRID

Many TCP protocol variants have demonstrated better performance in simulation and several limited network experiments. However, practical use in real applications of these protocols is still very limited because of the implementation and installation difficulties. On the other hand, users who need to transfer bulk data (e.g., ingrid/cloud computing) usually turn to application level solutions where these variants do not fair well. Among protocols considered in the application level solutions are UDP-based protocols, such as UDT (UDP-based DataTransport Protocol). UDT is considered one of the most recently developed new transport protocols with congestion control algorithms. It was developed to support next generation high-speed networks, including wide area optical networks. It is considered a state-of-the-art protocol, which promptly addresses various infrastructure requirements for transmitting data in high-speed networks. Its development, however, creates new vulnerabilities because like many other protocols, it is designed to rely solely on existing security mechanisms for existing protocols such as the Transmission Control Protocol (TCP) and User Datagram Protocol (UDP). Some of these security mechanisms cannot be used to absolutely protect UDT, just as security mechanisms devised for wired networks cannot be used to protect unwired ones. Both the recently developed UDT and the decades-old TCP/UDP lack a well-thought-out security architecture that addresses problems in today'snetworks. In this paper, we extend our previous work and implement our developed security methodology that can assist network and security investigators, designers, and users who consider and incorporate security when implementing UDT across wide area networks. These can support security architectural designs of UDP-based protocols as well as assist in the future development of other state-of-the-art fast data transfer protocols.