Abstract: A decentralized SLS monitoring structure (30) includes a measurement preparation and minimization module (32), measurement launching coordination module (34), measurement scheduler module (36) and measurement initialization and results module (38). The monitoring structure (30) prepares throughput tests to be performed on measurement engines (50), including a tapper (52) for splitting the data flow, packet collector (54) for time stamping packet headers, packet filter (56) for filtering packets according to a flow definition and packet recorder (58) for providing information to the measurement initialization and results module (38) for computing throughput.

Abstract: An architecture for measurement of per-hop, one-way delay includes a input observation circuit (24) at the input interface of a node (12) and a output observation circuit (26) at the output interface (16) of a node (12). The input observation circuit (24) copies and time stamps the headers of incoming packets, and filters packets according to an aggregate definition. Similarly, the output observation circuit (26) copies and time stamps the headers of outgoing packets, and filters packets according to the aggregate definition. The incoming and outgoing traces are correlated to calculate a delay measurement. A packet loss measurement uses input observation circuits (72a and 72b) at the input interfaces (14) of upstream and downstream nodes (20 and 22) and an output observation circuit 74 at the output interface (16) of the upstream node. The observation circuits (72a, 72b and 74) determine the number of lost packets across a node and between nodes, according to an aggregate definition.

Abstract: A decentralized SLS monitoring structure (30) includes a measurement preparation and minimization module (32), measurement launching coordination module (34), measurement scheduler module (36) and measurement initialization module (38). A configuration server (40) provides information regarding the test parameters for a given test and an SLS Objects database (42) provides information on the SLS specification for a given SLA (service level agreement).

Abstract: Apparatus and method for supporting operations and maintenance (“OAM”) functionality in an OBS network is described. One embodiment is an OBS network having OAM functionality, the OBS network comprising a plurality of OBS nodes interconnected via links. Each of the OBS nodes comprises an OAM module (“OAMM”) for processing information regarding OAM functions in the OBS network and a routing manager (“RM”) for processing routing information; and wherein at least one wavelength in each of the links comprises an OAM/1 wavelength for OAM/1 transmissions between nodes, the OAM/1 transmissions comprising OAM activity information; wherein at least one wavelength in each of the links comprises a reference wavelength for providing a wavelength reference to which light generating devices in the OBS network lock; and wherein at least one wavelength in each of the links comprises a routing wavelength for providing routing information between nodes.

Abstract: Apparatus and method for supporting operations and maintenance (“OAM”) functionality in an OBS network is described. One embodiment is an OBS network having OAM functionality, the OBS network comprising a plurality of OBS nodes interconnected via links. Each of the OBS nodes comprises an OAM module (“OAMM”) for processing information regarding OAM functions in the OBS network and a routing manager (“RM”) for processing routing information; and wherein at least one wavelength in each of the links comprises an OAM/1 wavelength for OAM/1 transmissions between nodes, the OAM/1 transmissions comprising OAM activity information; wherein at least one wavelength in each of the links comprises a reference wavelength for providing a wavelength reference to which light generating devices in the OBS network lock; and wherein at least one wavelength in each of the links comprises a routing wavelength for providing routing information between nodes.

Abstract: A partially preemptive quality of service (“QoS”) scheduling model for application in an optical burst switched (“OBS”) network node is described. In one aspect, the invention comprises a method for dropping only an portion of a data burst under consideration that overlaps with a portion of an already-scheduled data burst. In another aspect, the invention comprises preempting only that portion of an already-scheduled data burst that overlaps with a portion of a data burst under consideration and scheduling the data burst under consideration in its entirety. In one embodiment, the decision whether to drop a portion of the data burst under consideration or preempt a portion of the already-scheduled data burst is based on the relative priorities of the bursts. Similarly, the decision whether to perform a partial or full drop or preempt is based on the amount of overlap between the bursts.

Abstract: Apparatus and method for scheduling transmission of data bursts in an OBS network are described. One embodiment is a router in an OBS network, the router comprising a grouper module for collecting a group of BHPs arriving during a BHP collecting period and classifying each BHP of the group of BHPs into corresponding baskets depending on the data channel scheduling window of its corresponding data burst; a classifier and channel assignment module for assigning data bursts to appropriate data channels; and a channel scheduler for constructing an interval graph representing data bursts corresponding to BHPs classified in a single one of the baskets having overlapping segments on each data channel as adjacent vertices and scheduling data bursts on each data channel based on a process that finds a maximum number of non-adjacent vertices on the interval graph.

Abstract: A decentralized SLS monitoring structure (30) includes a measurement preparation and minimization module (32), measurement launching coordination module (34), measurement scheduler module (36) and measurement initialization and results module (38). The monitoring structure (30) prepares throughput tests to be performed on measurement engines (50), including a tapper (52) for splitting the data flow, packet collector (54) for time stamping packet headers, packet filter (56) for filtering packets according to a flow definition and packet recorder (58) for providing information to the measurement initialization and results module (38) for computing throughput.

Abstract: An architecture for measurement of per-hop, one-way delay includes a input observation circuit (24) at the input interface of a node (12) and a output observation circuit (26) at the output interface (16) of a node (12). The input observation circuit (24) copies and time stamps the headers of incoming packets, and filters packets according to an aggregate definition. Similarly, the output observation circuit (26) copies and time stamps the headers of outgoing packets, and filters packets according to the aggregate definition. The incoming and outgoing traces are correlated to calculate a delay measurement. A packet loss measurement uses input observation circuits (72a and 72b) at the input interfaces (14) of upstream and downstream nodes (20 and 22) and an output observation circuit 74 at the output interface (16) of the upstream node. The observation circuits (72a, 72b and 74) determine the number of lost packets across a node and between nodes, according to an aggregate definition.

Abstract: A decentralized SLS monitoring structure (30) includes a measurement preparation and minimization module (32), measurement launching coordination module (34), measurement scheduler module (36) and measurement initialization module (38). A configuration server (40) provides information regarding the test parameters for a given test and an SLS Objects database (42) provides information on the SLS specification for a given SLA (service level agreement).

Abstract: Apparatus and method for scheduling transmission of data bursts in an OBS network are described. One embodiment is a router in an OBS network, the router comprising means for collecting a group of BHPs arriving during a BHP collecting period and classifying each BHP of the group of BHPs into corresponding baskets depending on the data channel scheduling window of its corresponding data burst; means for assigning data bursts to appropriate data channels; and means for constructing an interval graph representing data bursts corresponding to BHPs classified in a single one of the baskets having overlapping segments on each data channel as adjacent vertices and scheduling data bursts on each data channel based on a process that finds a maximum number of non-adjacent vertices on the interval graph.

Abstract: A partially preemptive quality of service (“QoS”) scheduling model for application in an optical burst switched (“OBS”) network node is described. In one aspect, the invention comprises a method for dropping only an portion of a data burst under consideration that overlaps with a portion of an already-scheduled data burst. In another aspect, the invention comprises preempting only that portion of an already-scheduled data burst that overlaps with a portion of a data burst under consideration and scheduling the data burst under consideration in its entirety. In one embodiment, the decision whether to drop a portion of the data burst under consideration or preempt a portion of the already-scheduled data burst is based on the relative priorities of the bursts. Similarly, the decision whether to perform a partial or full drop or preempt is based on the amount of overlap between the bursts.

Abstract: A load sharing system which minimizes overall costs by assigning segments of a divisible load to distributed processor platforms based on the resource utilization cost of each processor platform. The distributed processor platforms are connected via data links which also have associated resource utilization costs. A controller divides a divisible load or task and assigns each segment of the load or task to a processor platform based on the processor platform's resource utilization cost and data link cost. After the initial allocation, an optimizing reallocation is performed to reduce the overall monetary cost processing the load or task. The optimization can be performed using a pair-wise swapping technique.