Abstract: A system and method are provided to route packets in a data center network. Individual packets are encapsulated at an edge of the data center network, so that each encapsulated packet includes a set of header fields, such as a tenant identifier. For each encapsulated packet, a hash class is determined from the set of header fields. A routing virtual local area network (VLAN) is selected for the packet based on the tenant identifier and the hash class.

Abstract: In a method for routing packets between a plurality of switches in a computer network, in which paths between the plurality of switches are identified as a plurality of virtual local area networks (VLANs) stored in a balancing table, a packet to be routed from a source switch to a destination switch is received. In addition, a VLAN is selected from the plurality of VLANs in the balancing table to route the packet through the computer network and the packet is routed through the selected VLAN.

Abstract: A system and method are provided to route packets in a data center network. Individual packets are encapsulated at an edge of the data center network, so that each encapsulated packet includes a set of header fields, such as a tenant identifier. For each encapsulated packet, a hash class is determined from the set of header fields. A routing virtual local area network (VLAN) is selected for the packet based on the tenant identifier and the hash class.

Abstract: Embodiments herein relate to addition or modification to a forwarding table based on an address. A first packet having a source address and a location value may be received. The source address includes a source of the first packet and the location value indicates at least part of a route along a network to the source address. The forwarding table is not modified or no new entry is added to the forwarding table, if the forwarding table does not include the source address.

Abstract: An optical interconnect (200) includes: a reflective body (230) having a first reflective surface (235) and a second reflective surface (240) opposite the first reflective surface (235); a first optical waveguide (205) that directs a first optical signal received from a first communicating device (105) to the first reflective surface (235); a second optical waveguide (210) that directs the first optical signal from the first reflective surface (235) of the reflective body (230) to a second communicating device (110); a third optical waveguide (215) that directs a second optical signal received from the second communicating device (110) to the second reflective surface (240) of the reflective body (230); and a fourth optical waveguide (220) that directs the second optical signal from the second reflective surface (240) of the reflective body (230) to the first communicating device (105).

Abstract: In one implementation, a cable harness switch includes a plurality of input ports, a first plurality of output ports, a second plurality of input ports, and a circuit switch module. Each input port from the plurality of input ports is configured to be coupled to a network link. Each output port from the first plurality of output ports is configured to be coupled to a network link. Each output port from the second plurality of output ports configured to be coupled to a network switch device. The circuit switch module is operatively coupled to the plurality of input ports, the first plurality of output ports, and the second plurality of output ports to define a network circuit including an input port from the plurality of input ports and an output port from the first plurality of output ports and the second plurality of output ports.

Abstract: An apparatus comprises a given multimode optical waveguide extending in a given direction. The apparatus also comprises another multimode optical waveguide extending in another direction and intersecting with the given multimode waveguide. The apparatus further comprises a bi-stable optical switch positioned at the intersection of the given multimode optical waveguide and the another multimode optical waveguide to redirect a multimode optical signal transmitted on the given multimode optical waveguide to the another optical waveguide in a redirection state and pass the multimode optical signal transmitted on the given multimode optical waveguide across the intersection of the given multimode optical waveguide and the another optical waveguide in a pass-through state. The bi-stable optical switch can comprise a gap extending diagonally from a given corner of the intersection of the given and the another optical multimode waveguides to an opposing corner of the intersection.

Abstract: According to one embodiment of the present invention, there is provided a system for allocating bandwidth in a network to a plurality of traffic classes. Each traffic class has a first bandwidth allocation. The system comprises a network manager which is configured to determine a bandwidth utilization for each traffic class, to determine an amount of unused network bandwidth, to calculate second bandwidth allocations for each traffic class by allocating a share of any determined unused network bandwidth between at least some of the traffic classes, and to update, in accordance with the second bandwidth allocations, a routing table accessible by routers in the network.

Abstract: According to one example of the present invention, there is provided a method of routing data packets to a plurality of packet processors in a computer network. The method comprising obtaining workload data from the packet processors, determining a workload distribution across the packet processors, and updating a balancing table used by a switching element in the network based on the determined workload.

Abstract: In a method (400) for routing packets between a plurality of top switches (110a-110n) and a plurality of leaf switches (120a-120n) using a balancing table (204, 208, 210) in a fat tree network (100), a failed link between at least one top switch (110n) and at least one leaf switch (120n) is detected (402). In addition, the balancing table (204, 208, 210) is modified (406) based on the detected failed link, and the packets are routed (408) between the plurality of top switches (110a-110n) and the plurality of leaf switches (120a-120n) in the fat tree network (100) based on the modified balancing table (204, 208, 210).

Abstract: A dynamic pinning remote direct memory access is performed by creating sections of data to be transferred through a remote direct memory access. Each section includes a subset of the data to be transferred or received. To perform the remote direct memory access, each section is pinned, used for the remote direct memory access, and released after the transfer is complete.

Abstract: Described is a switch architecture that combines address management with simplified hardware to implement fast route lookup within network switches such as Ethernet switches. A managed address includes a cluster ID which is shared by all endpoints in a cluster, and a member ID which is unique for each node in the cluster. The switch extracts the cluster ID from a target address for a packet and compares it against at least one cluster ID stored in a cluster identification memory. Responsive to a match, the switch generates a port identification for the packet using a fast lookup table. Responsive to no match, the target address is considered an unmanaged address. In one implementation, a slow lookup table can be used to generate a port identification for the unmanaged address.

Abstract: According to one example of the present invention, there is provided a method of routing data packets to a plurality of packet processors in a computer network. The method comprising obtaining workload data from the packet processors, determining a workload distribution across the packet processors, and updating a balancing table used by a switching element in the network based on the determined workload.

Abstract: A method for load balancing Ethernet traffic within a fat tree network (315, 455) includes randomly assigning incoming messages (510) into hash classes using a hash function (520); allocating the hash classes among uplinks (550); and transmitting the incoming messages on the uplinks (550) according to the hash class. A network switch (515) for load balancing communication flows in a fat tree network (315, 455) includes downlinks (545) and uplinks (550); the network switch (515) being configured to route communication flows among the downlinks (545) and uplinks (550); a hash module (520) which receives a MAC address from a message (510) and outputs a hash address; and a TCAM lookup module (535) which allocates the hash address into a hash class and allocates the hash class to one of the uplinks (550).

Abstract: In a method for routing packets between a plurality of switches in a computer network, in which paths between the plurality of switches are identified as a plurality of virtual local area networks (VLANs) stored in a balancing table, a packet to be routed from a source switch to a destination switch is received. In addition, a VLAN is selected from the plurality of VLANs in the balancing table to route the packet through the computer network and the packet is routed through the selected VLAN.

Abstract: According to one embodiment of the present invention, there is provided a system for allocating bandwidth in a network to a plurality of traffic classes. Each traffic class has a first bandwidth allocation. The system comprises a network manager which is configured to determine a bandwidth utilization for each traffic class, to determine an amount of unused network bandwidth, to calculate second bandwidth allocations for each traffic class by allocating a share of any determined unused network bandwidth between at least some of the traffic classes, and to update, in accordance with the second bandwidth allocations, a routing table accessible by routers in the network.

Abstract: A description of components in a reconfigurable data center is received. A set of components from the description is selected for a physical platform based on a logical platform specification.

Abstract: Methods and systems for performing direct memory access (DMA) transfers are described. An invalidate queue (or other storage device) contains an entry associated with a DMA transfer in progress. If the invalidate queue detects an invalidation of a memory page associated with that entry, then it is marked invalid. If the entry is marked invalid during the DMA transfer, then that DMA transfer is aborted. This enables, among other things, DMA transfers to unpinned virtual memory.

Abstract: Methods and systems for caching data from a head end of a queue are described. The cached data can then be selectively forwarded from the data producer to the data consumer upon request.

Abstract: The present invention is a method of and system for program thread synchronization. In accordance with an embodiment of the invention, a method of synchronizing program threads for one or more processors is provided. An instruction cache line for is determined each of a plurality of program threads to be synchronized. For each processor executing one or more of the threads to be synchronized, execution of the thread is halted at a barrier by rendering the determined instruction cache line unavailable. Execution of the threads resumes by rendering the determined instruction cache lines available.