Abstract: The performance of randomized load balanced or selective, randomized load balanced networks is enhanced by using ingress traffic engineering in addition to randomized traffic splitting. By first using the capacity of all links leading to the final destination of traffic, the remaining capacity is freed up for best effort traffic. Traffic splitting rules that enhance the performance of randomized load balanced networks in terms of packet missequencing and other quality of service criteria are also described.

Abstract: Described is a technology by which an assignment model is computed to distribute labeling tasks among judging entities (judges). The assignment model is optimized by obtaining accuracy-related data of the judges, e.g., by probing the judges with labeling tasks having a gold standard label and evaluating the judges' labels against the gold standard labels, and optimizing for accuracy. Optimization may be based upon on or more other constraints, such as per-judge cost and/or quota.

Abstract: The performance of randomized load balanced or selective, randomized load balanced networks is enhanced by using ingress traffic engineering in addition to randomized traffic splitting. By first using the capacity of all links leading to the final destination of traffic, the remaining capacity is freed up for best effort traffic. Traffic splitting rules that enhance the performance of randomized load balanced networks in terms of packet missequencing and other quality of service criteria are also described.

Abstract: Broadly, techniques for solving network routing within a predetermined error are disclosed. These techniques may be applied to networks supporting dedicated reserve capacity, where reserved capacity on links in the network is dedicated for a particular commodity (generally, a source and sink pair of computers), and shared recovery, where reserved capacity on links is shared amongst two or more commodities. These techniques use an iterative process to determine flows on each of the links in a network. Costs are set for each commodity, and primary and secondary (i.e., backup) flows are initialized. A commodity is selected and demand for the commodity is routed through the shortest path. Costs are updated for each potential failure mode. For each commodity, the flows and costs are updated. Once all flows and costs are updated, then it is determined if a function is less than a predetermined value.

Abstract: A method for exchanging routing information between I-BGP routers within an autonomous system (AS) advantageously enables a solution to both persistent route oscillation problems and transient route oscillation problems which may occur when using I-BGP in a given AS. Conventional I-BGP protocol techniques are extended by enabling I-BGP speakers (e.g., routers) to communicate a set of possible best paths to a given destination, rather than communicating only a single best path, to each of their I-BGP peers within the given AS. Specifically, a plurality of possible best paths to a destination are communicated (where there are in fact more than one) from an I-BGP speaker in a given AS to its I-BGP peers (within the given AS), for each neighboring AS that provides any such paths (i.e., routes to the destination).

Abstract: Broadly, techniques for solving network routing within a predetermined error are disclosed. These techniques may be applied to networks supporting dedicated reserve capacity, where reserved capacity on links in the network is dedicated for a particular commodity (generally, a source and sink pair of computers), and shared recovery, where reserved capacity on links is shared amongst two or more commodities. These techniques use an iterative process to determine flows on each of the links in a network. Costs are set for each commodity, and primary and secondary (i.e., backup) flows are initialized. A commodity is selected and demand for the commodity is routed through the shortest path. Costs are updated for each potential failure mode. For each commodity, the flows and costs are updated. Once all flows and costs are updated, then it is determined if a function is less than a predetermined value.