Off-campus UMass Amherst users: To download campus access theses, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users: Please talk to your librarian about requesting this thesis through interlibrary loan.
Theses that have an embargo placed on them will not be available to anyone until the embargo expires.
Electrical & Computer Engineering
Master of Science in Electrical and Computer Engineering (M.S.E.C.E.)
Year Degree Awarded
Month Degree Awarded
Packet Pacing, Small-Buffer, QLBP, all optical core, pacer
The growing use of the internet and the wide variety of applications which run on it puts a considerable demand for high bandwidth networks. All optical core networks are one such possible networks which cater to the demand of high bandwidths.Since the all optical routers use the fiber delay lines as optical buffers, it is impossible to build optical buffers of such high capacity.The present day solutions for optical buffers are fiber delay lines(FDL) which are nothing but long optical fiber lines which are convoluted and folded in order to provide the necessary delay in transmission resulting in a small buffer which can store packets and thus can be used as a buffer.If we consider the example of a single TCP source sending an infinite amount of data with packets of constant size with the flow passing through a single router. If we make an assumption that the sender's access link is much faster than the receiver's bottleneck link of capacity, it will cause packets to be queued at the router.We propose a mechanism to pace traffic in the network based on the queue length of the buffer in the output port. The underlying principle delays the transmission of the packet depending on the instantaneous queue length of the buffer.A prototype of such a model was simulated in network simulator and the performance metrics were measured.