Abstract:
Disclosed are various embodiments for improving hash table utilization. A key corresponding to a data item to be inserted into a hash table can be transformed to improve the entropy of the key space and the resultant hash codes that can generated. Transformation data can be inserted into the key in various ways, which can result in a greater degree of variance in the resultant hash code calculated based upon the transformed key.
Abstract:
Methods and apparatus for load balancing data traffic are disclosed. An example method includes determining a respective quality metric for each of a plurality of members of an aggregation group of the network device, each respective quality metric representing respective data traffic loading for each member of the plurality of aggregation group members. The example method also includes grouping the plurality of aggregation members into a plurality of loading/quality bands based on their respective quality metric. The example method further includes selecting members of the aggregation group for transmitting packets from a loading/quality band corresponding with members of the aggregation group having lower data traffic loading relative to other members of the aggregation group.
Abstract:
Network devices facilitate flow management through packet marking. The network devices may be switches, routers, bridges, hubs, or any other network device. The packet marking may include analyzing received packets to determine when the received packets meet a marking criterion, and then applying a configurable marking function to mark the packets in a particular way. The marking capability may facilitate deadline aware end-to-end flow management, as one specific example. More generally, the marking capability may facilitate traffic management actions such as visibility actions and flow management actions.
Abstract:
A method for diagnosing performance of a network switch device includes a processor monitoring data generated by a sensor associated with a network switch device, the data related to states or attributes of the network switch device. The processor detects a determined condition in the operation of the network switch device related to the state or attribute. The processor generates an event trigger in response to detecting the determined condition and executes a forensic command in response to the event trigger. Executing the command includes sending information relevant to the determined condition for aggregation in computer storage and for analysis.
Abstract:
Network devices perform multiple stage path resolution. The path resolution may be ECMP resolution. Any particular stage of the multiple stage path resolution may be skipped under certain conditions. Further, the network device facilitate redistribution of traffic when a next hop goes down in a fast, efficient manner, and without reassigning traffic that was going to other unaffected next hops, using multiple stage ECMP resolution.
Abstract:
In various embodiments, a system includes a switch comprising a resource that is shared between multiple objects. The switch comprises circuitry that determines a congestion metric for the switch in response to an amount of used of the resource by the objects. The circuitry determines a feedback parameter that is responsive to the congestion metric. The circuitry generates a congestion notification message that comprises a congestion feedback value responsive to the feedback parameter. In further embodiments, a system includes a switch that processes data for a first data link layer access network. The switch includes circuitry that identifies whether a received packet originated from a source device that shares the same network layer access layer as the switch. If the source device shares the same network layer access network as the switch, the circuitry generates a congestion notification message comprising a congestion feedback value for the switch.
Abstract:
A system for multicast switching for distributed devices may include an ingress node including an ingress memory and an egress node including an egress memory, where the ingress node is communicatively coupled to the egress node. The ingress node may be operable to receive a portion of a multicast frame over an ingress port, bypass the ingress memory and provide the portion to the egress node when the portion satisfies an ingress criteria, otherwise receive and store the entire frame in the ingress memory before providing the frame to the egress node. The egress node may be operable to receive the portion from the ingress node, bypass the egress memory for the portion and provide the portion to the first egress port when an egress criteria is satisfied, otherwise receive and store the entire multicast frame in the egress memory before providing the multicast frame to an egress port.
Abstract:
In various embodiments, a system includes a switch comprising a resource that is shared between multiple objects. The switch comprises circuitry that determines a congestion metric for the switch in response to an amount of used of the resource by the objects. The circuitry determines a feedback parameter that is responsive to the congestion metric. The circuitry generates a congestion notification message that comprises a congestion feedback value responsive to the feedback parameter. In further embodiments, a system includes a switch that processes data for a first data link layer access network. The switch includes circuitry that identifies whether a received packet originated from a source device that shares the same network layer access layer as the switch. If the source device shares the same network layer access network as the switch, the circuitry generates a congestion notification message comprising a congestion feedback value for the switch.
Abstract:
Disclosed are various embodiments that relate identifying a source of corruption in a network made up of multiple network nodes. A network node is configured to provide corruption source identification while handling packets according to a cut-through scheme. According to some embodiments, a network node may perform a running error detection operation on a cut-through packet and then insert a debug indicator into the cut through packet. In other embodiments, the network node may process some packets according to a cut-through scheme while process other packets according to a store-and-forward scheme to detect packet corruption in a network.
Abstract:
Disclosed are various embodiments for improving hash table utilization. A key corresponding to a data item to be inserted into a hash table can be transformed to improve the entropy of the key space and the resultant hash codes that can generated. Transformation data can be inserted into the key in various ways, which can result in a greater degree of variance in the resultant hash code calculated based upon the transformed key.