A method to determine a hierarchical fair share of resources among one or more scheduling peers. The method determines a greatest fairness difference between the usage percentage difference of a most-served competing peer and that of a least-served competing peer. The method determines a greatest accumulated fairness difference between the accumulated usage percentage difference of the most-served competing peer and that of the least-served competing peer. The method adjusts a resource allocation such that resources from the most-served competing peer are allocated to the least-served competing peer if an adjustment condition is met, wherein the adjustment condition is based on one or both of i) the greatest fairness difference, and ii) whether the greatest accumulated fairness difference exceeds at least one threshold.

A system for providing communications over a communications network includes a communications interface and a processor. The communications interface communicates over the communications network. The processor directs a communications scheduler to determine at least one metric for a path within the communications network. The processor also selects a data flow for the path and determines whether to transmit a packet in the selected data flow based on the at least one metric. The processor then directs a communications protocol handler to generate the packet for the selected data flow.

In one embodiment, primary circuits may be established on a network link, where each primary circuit individually allocates primary bandwidth of the network link and is assigned to one of either a respective primary weighted queue of the network link corresponding to a size of the respective primary circuit or a shared primary weighted queue of the network link corresponding to a total size of the individually allocated primary bandwidth for the network link. In addition, protection circuits may also be established on the network link, where the protection circuits collectively allocate shared backup bandwidth of the network link and are assigned to a shared backup weighted queue of the network link corresponding to a size of the shared backup bandwidth for the network link. As such, primary and backup packets may be forwarded on the network link according to the primary and backup weighted queues (e.g., allowing 1:1 protection).

Scheduling of packets to be forwarded onto DOCSIS downstream channels as part of a Virtual Converged Cable Access Platform (CCAP). A packet to be forwarded onto a DOCSIS downstream channel is enqueued in a service flow queue. The packets stored in the service flow queue are associated with a single service flow. A request is propagated up a hierarchy of schedule elements to a scheduler process to schedule the packet for delivery. The scheduler process determines a grant of how much traffic to offer the DOCSIS downstream channel. The grant determined for the DOCSIS downstream channel may be expressed in units of symbols rather than in bytes. The scheduler process extends a particular grant to the service flow queue by translating symbols in the grant for the service flow queue which issued the request.

Methods, apparatuses, and computer-readable medium for providing a fairness protocol in a network element are disclosed herein. An example method includes receiving one or more packets at each of a plurality of ingress ports of the network element, and scheduling the packets into a plurality of queues, wherein each of the queues is associated with packets that are sourced from one of the ingress ports. The method also includes monitoring a bandwidth of traffic sourced from each of the ingress ports, identifying a port among the ingress ports that sources a smallest bandwidth of traffic, and arbitrating among the queues when transmitting packets from an egress port of the network element by giving precedence to the identified port that sources the smallest bandwidth of traffic. Additionally, arbitrating among the queues distributes a bandwidth of the egress port equally among the ingress ports.

A system for providing communications over a communications network includes a communications interface and a processor. The communications interface communicates over the communications network. The processor directs a communications scheduler to determine at least one metric for a path within the communications network. The processor also selects a data flow for the path and determines whether to transmit a packet in the selected data flow based on the at least one metric. The processor then directs a communications protocol handler to generate the packet for the selected data flow.

A method to determine a hierarchical fair share of resources among one or more scheduling peers. The method determines a greatest fairness difference between the usage percentage difference of a most-served competing peer and that of a least-served competing peer. The method determines a greatest accumulated fairness difference between the accumulated usage percentage difference of the most-served competing peer and that of the least-served competing peer. The method adjusts a resource allocation such that resources from the most-served competing peer are allocated to the least-served competing peer if an adjustment condition is met, wherein the adjustment condition is based on one or both of i) the greatest fairness difference, and ii) whether the greatest accumulated fairness difference exceeds at least one threshold.

A compute node with multiple transfer processes that share an Infiniband connection to send and receive messages across a network. Transfer processes are first associated with an Infiniband queue pair (QP) connection. Then send message commands associated with a transfer process are issued. This causes an Infiniband message to be generated and sent, via the QP connection, to a remote compute node corresponding to the QP. Send message commands associated with another process are also issued. This causes another Infiniband message to be generated and sent, via the same QP connection, to the same remote compute node. As mentioned, multiple processes may receive network messages received via a shared QP connection. A transfer process on a receiving compute node receives a network message through a QP connection using a receive queue. A second transfer process receives another message through the same QP connection using another receive queue.

A technique allows stations to utilize an equal share of resources (e.g., airtime or throughput). This prevents slow stations from consuming too many resources (e.g., using up too much air time). Fairness is ensured by selective dropping after a multicast packet is converted to unicast. This prevents slow stations from using more than their share of buffer resources. Multicast conversion aware back-pressure into the network layer can be used to prevent unnecessary dropping of packets after multicast to unicast (1:n) conversion by considering duplicated transmit buffers. This technique helps achieve airtime/resource fairness among stations.