Abstract: A method for scheduling a crossbar using distributed request-grant-accept arbitration between input group arbiters and output group arbiters in a switch unit is provided. The switch unit may be a hierarchical high radix switch with a timesliced crossbar that is configured to transfer packets between a plurality of input ports and a plurality of output ports, organized into groups, using wide words. The timesliced crossbar transfers data for a given packet once per supercycle, in a designated timeslice of that supercycle. Multiple buffered packets from one input port to multiple output ports are transferred by utilizing different timeslices of the supercycle.

Abstract: A method for scheduling a crossbar using distributed request-grant-accept arbitration between input group arbiters and output group arbiters in a switch unit is provided. The switch unit may be a hierarchical high radix switch with a timesliced crossbar that is configured to transfer packets between a plurality of input ports and a plurality of output ports, organized into groups, using wide words. The timesliced crossbar transfers data for a given packet once per supercycle, in a designated timeslice of that supercycle. Multiple buffered packets from one input port to multiple output ports are transferred by utilizing different timeslices of the supercycle.

Abstract: Congestion management for data traffic in a virtual domain identifies a congestion source and sends a message to the source to adjust data traffic rates. The source may be a virtual machine hosted by a physical server with one or more virtual servers incorporated. A congestion manager may identify the source and send the message to the source without affecting other data sources hosted by the physical server or the virtual servers. In some embodiments, information about the congestion source may be encapsulated in a packet payload readable only by the congestion source so only the congestion source receives the instruction to adjust the transmission rate.

Abstract: In one embodiment, a system includes an initiator port adapted to receive a traffic flow having a plurality of packets and a hardware processor and logic integrated with and/or executable by the processor, the logic being adapted to classify the traffic flow into a traffic class based on at least one criteria related to the traffic flow, the traffic class being selected from a plurality of traffic classes and transmit one or more packets of the traffic flow to a target port designated by the one or more packets of the traffic flow until an amount of flow credits allocated to the target port is insufficient to service additional packets. Other systems, methods, and computer program products for providing credit-based flow control are described in more embodiments.

Abstract: A method for transmitting data packets from a first node to a second node. The method includes transmitting the data packet from the first node to the one second node where each data packet is determined to be sent according to a first or a second transmission mode, where, in the first transmission mode, the data packet is transmitted to the second node according to a prescheduled scheme, where the prescheduled scheme defines a cyclic one-to-one assignment between first and second node over time slots so that the data packet is forwarded to the second node during that time slot the one-to-one assignment of which assigns the first node with the one second node and in the second transmission mode, the data packet is transmitted during an actual time slot to the second node while overruling the prescheduled scheme.

Abstract: Methods and apparatus are provided for routing data packets between source and destination switches in a fat tree network. For each packet, a route is selected having three or less routing phases such that the route follows a shortest path across the network between the source and destination switches. The data packet is transmitted from the source switch to the destination switch, via the route, on one of first and second virtual channels unless the route includes a predetermined one of a down-up turn and an up-down turn. If the route includes the predetermined one of a down-up turn and an up-down turn, the data packet is transmitted, via the route, on the first virtual channel up to the switch 1 at which the turn occurs and on the second virtual channel from that switch. This provides full-connectivity in fat tree networks with deadlock free operation.

Abstract: A virtual network is implemented on a physical network. A virtual network data packet is tunneled through the physical network via encapsulation within a physical network data packet and via transmission of the physical network data packet through the physical network. A network congestion notification capability of the virtual network is preserved and modified during transmission of virtual network data through the physical network and vice-versa. Congestion notification metadata can be copied from a header of a virtual network data packet to a header of a physical network data packet when the virtual network data packet is encapsulated into the physical network data packet. Congestion notification metadata can be copied from a header of a physical network data packet to a header of a virtual network data packet when the virtual network data packet is decapsulated from the physical network data packet.

Abstract: A system for hierarchical occupancy based congestion management includes a buffer embodied in a computer readable storage medium including a plurality of buffer units for storing packets of a data flow received from sources. The system includes a buffer manager that stores information about the packets stored in the buffer, including a selection criterion associated with each of the plurality of sources and a congestion estimator that monitors a congestion level in the buffer. The system also includes a occupancy sampler that randomly selects at least two occupied buffer units from the plurality of buffer units and identifies the source of the packet stored in each of the occupied buffer units and a congestion notification message generator that generates a congestion notification message; wherein if the congestion level in the buffer exceeds a threshold value the congestion notification message is sent to the identified source with a higher selection criteria.

Abstract: A method for hierarchical occupancy based congestion management includes receiving, by a computing device, a plurality of data flows, each of the plurality of data flows is received from a source and includes a plurality of data packets and storing the plurality of data packets in a buffer including multiple storage units. The method includes determining if the buffer is congested, responsive to determining the buffer is congested randomly selecting at least two occupied units of the buffer and identifying a source of each of the data packets stored in the occupied units of the buffer and generating a congestion notification message. The method also includes comparing a selection criterion associated with each identified source to determine which identified source has a higher selection criterion and sending the congestion notification message to the identified source with the higher selection criterion.

Abstract: An approach is described for routing data to a plurality of output terminals via a integrated switch router including a crossbar switch having both a crossbar and a plurality of crossbar bypass lines. Whereas the crossbar may connect each input of the crossbar switch to each output of the crossbar switch, each of the plurality of crossbar bypass lines may connect a single input of the crossbar switch to a corresponding single output of the crossbar switch. According to such approach, a replicated copy of a multicast packet may be forwarded to an output terminal via a crossbar bypass line in parallel with other data forwarded via the crossbar, thus increasing integrated switch router bandwidth.

Abstract: An approach is described for routing data to a plurality of output terminals via a integrated switch router including a crossbar switch having both a crossbar and a plurality of crossbar bypass lines. Whereas the crossbar may connect each input of the crossbar switch to each output of the crossbar switch, each of the plurality of crossbar bypass lines may connect a single input of the crossbar switch to a corresponding single output of the crossbar switch. According to such approach, a replicated copy of a multicast packet may be forwarded to an output terminal via a crossbar bypass line in parallel with other data forwarded via the crossbar, thus increasing integrated switch router bandwidth.

Abstract: A network fabric may divide a physical connection into a plurality of VLANs as defined by IEEE 802.1Q. Moreover, many network fabrics use Priority Flow Control to identify and segregate network traffic based on different traffic classes or priorities. Current routing protocols define only eight traffic classes. In contrast, a network fabric may contain thousands of unique VLANs. When network congestion occurs, network devices (e.g., switches, bridges, routers, servers, etc.) can negotiate to pause the network traffic associated with one of the different traffic classes. Pausing the data packets associated with a single traffic class may also stop the data packets associated with thousands of VLANs. The embodiments disclosed herein permit a network fabric to individually pause VLANs rather than entire traffic classes.

Abstract: According to one embodiment, a method for handling congestion in a network includes determining that there is congestion on a first device in a network, setting a congestion indicator in a header of a packet to indicate an amount of congestion at the first device, sending the packet to all devices that send traffic to the first device, receiving the packet having the multi-bit indicator in a header thereof at a device that sends traffic to the first device, and reducing a congestion window by a factor of between about 5% and about 50% based on a severity of the congestion indicated by the multi-bit indicator, wherein the congestion window is reduced by a greater factor when the congestion is indicated as being more severe. Other systems and methods for handling congestion in a network are described according to more embodiments.

Abstract: Embodiments of the invention relate to providing virtual network domain movement operations for overlay networks. One embodiment includes a method that includes determining one or more overlay network attributes (ONAs) for a plurality of virtual networks. The one or more ONAs are associated with the virtual networks. The one or more ONAs are managed as one or more portable entities by one or more of creating ONAs, deleting ONAs, moving ONAs, combining ONAs and dividing ONAs. A movement operation is performed on the one or more virtual networks among one or more servers of one or more overlay networks based on the management of the one or more ONAs.

Abstract: A method for hybrid arrival-occupancy based congestion management includes increasing a recent arrivals counter associated with a data flow from one or more sources in response to receiving a data packet from one of the sources and storing the data packet in a buffer including multiple storage units. The method includes determining if a buffer is congested, randomly selecting an occupied unit of the buffer and determining the source of the packet stored in the occupied unit of the buffer, generating a congestion notification message, sending the congestion notification message to the source of the packet stored in the occupied unit of the buffer if the recent arrivals counter exceeds a threshold value and decreasing the recent arrivals counter associated with the source of the packet stored in the occupied unit of the buffer and discarding the congestion notification message if the recent arrivals counter has a zero value.

Abstract: Embodiments relate to bypassing congestion points in a network. An aspect includes sampling queues of a plurality of switches in a network. When packet congestion is detected at a congestion point of a first switch, the packet flow contributing to the packet congestion is identified. A congestion notification message indicating the identified packet flow is then propagated to upstream switches, which are upstream from the first switch in the network. The congestion notification message is then snooped by the upstream switches. Virtual queues within the upstream switches are associated with the identified packet flow to hold packets associated with the identified packet flow. The packets associated with the identified packet flow are then re-routed to bypass the packet congestion in the first switch.

Abstract: Techniques are disclosed to implement a scheduling scheme for a crossbar scheduler that provides distributed request-grant-accept arbitration between input group arbiters and output group arbiters in a distributed switch. Input and output ports are grouped and assigned a respective arbiter. The input group arbiters communicate requests indicating a count of respective ports having data packets to be transmitted via one of the output ports. The output group arbiter attempts to accommodate the requests for each member of an input group before proceeding to a next input group.

Abstract: A method and system are provided for implementing a hierarchical high radix switch with a time-sliced crossbar. The hierarchical high radix switch includes a plurality of inputs and a plurality of outputs. Each input belongs to one input group; each input group sends consolidated requests to each output, by ORing the requests from the local input ports in that input group. Each output port belongs to one output group; each output port grants one of the requesting input groups using a rotating priority defined by a next-to-serve pointer. Each output group consolidates the output port grants and allows one grant to pass back to an input group. Each input port in an input group evaluates all incoming grants in an oldest packet first manner to form an accept. Each input group consolidates the input port accepts and selects one accept to send to the output port.

Abstract: Embodiments relate to controlling workload flow on converged Ethernet links. An aspect includes coupling, by a processing device, a first control loop to a second control loop. The second control loop monitors the operation of the first control loop. An equilibrium set point is initialized for the second control loop prior to commencing operation of the first control loop. Accordingly, the equilibrium set point value is adjusted in the second control loop continuously based on a rate of operation of the first control loop.

Abstract: Embodiments relate to proactively probing the packet queues of elements in a physical or virtual network to predict and prevent the occurrence of congestion points. An aspect includes receiving a first feedback request at a central controller connected to a plurality of switches in a network. The first feedback request includes a request to periodically probe a status of queues of switches in the network. A second feedback request is then transmitted to one or all the switches in a path leading to a designated destination. Responses to the second feedback request are received at the central controller from a designated proxy switch, which aggregated the responses into a single data packet. Accordingly, the responses extracted from the single data packet at the central controller are used to preventing future congestion points.