Browsing by Author "Eom, Hyeonsang"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item Improving Link-State Routing - by Using Estimated Future Link Delays (Revised)(2003-08-01) Eom, HyeonsangIn link-state routing, routes are determined based on estimates of the current delays on the links. Ideally, a data packet should be routed based on the delays it will encounter at each link of the path at the time the packet gets to the link. To address this issue, we have developed a new approach that improves link-state routing by estimating and using the future link delays encountered by data packets. In link-state routing, link-delay estimates are periodically flooded throughout the network. This flooding of link-delay estimates is done without considering the relevance of these estimates to routing quality. Our approach also improves link-state routing by broadcasting these estimates only to the extent that they are relevant. Simulation studies suggest that our approach can lead to significant reductions in routing traffic with noticeable improvements of routing quality in high-load conditions. (UMIACS-TR-2001-75.2)Item Information Dynamics Applied to Link-State Routing(2002-01-31) Eom, Hyeonsang; Agrawala, Ashok K.; Noh, Sam H.; Shankar, A. UdayaInformation Dynamics is an information-centric framework that provides a sufficient understanding of the characteristics of information used in systems for better system design and implementation. In this paper, we describe how to improve link-state routing based on this framework. Link-state routing protocols such as OSPF (Open Shortest Path First) are currently used in many networks. In link-state routing, routes are determined based on link-delay estimates, which are periodically flooded throughout the network. This flooding of link-delay estimates is done without considering the relevance of these estimates to routing quality, i.e. without taking into account the usefulness of the link-delay information. We have developed a new approach that improves link-state routing by estimating future link delays and flooding these estimates only to the extent that they are relevant. This means that we consider the dynamics of the link-delay information and its usefulness. Simulation studies suggest that our approach can lead to significant reductions in routing traffic with noticeable improvements of routing quality in high-load conditions, demonstrating the effectiveness of the framework. We plan to further investigate the conditions where our information-dynamics approach is better than the standard approach. (Also UMIACS-TR-2001-75)Item Load Balancing Factor (LBF): A Workload Migration Metric(2003-08-01) Eom, Hyeonsang; Hollingsworth, Jeffrey K.We introduce a new performance metric, called Load Balancing Factor (LBF), to evaluate different tuning alternatives of workload migration within a distributed/parallel program. The metric is unique because it shows the performance implications of a specific tuning alternative rather than quantifying where time is spent in the program. Previously we developed a variation of the metric for coarse-grained process placement, and demonstrated that it accurately predicts the placement impact. In this paper we focus on a variation designed for fine-grained function shipping in a client/server environment and present its online algorithm. We use a synthetic application to show that LBF provides accurate guidance about procedure-level migration. (UMIACS-TR-2002-55)