Future Wei Technologies, Inc.

Serhat Nazim Avci

Fangping Liu

Cedric Westphal

Zhenjiang Li

Xiaogang Zhu

Methods and apparatus for determining and adjusting traffic rate limits to reduce queuing delays and latency in a data center network. In various embodiments, a network controller receives network topology information and traffic data relating to flows of the network. The traffic data includes traffic matrix information, flow path information, and pause frame distribution information. Based on this information, the network controller identifies server network node interfaces that are the source of a pause frame. Flows associated with such network node interfaces are also identified and ranked based on traffic volume. For ranked flows meeting predetermined criteria, reduced traffic rate limits are calculated and sent to associated server network node interfaces. In some embodiments, the reduced traffic rate limits are incrementally recovered when normal traffic resumes on a paused link. In further embodiments, traffic rate limits are also adjusted for flows of highly utilized and underutilized network node interfaces.

Method and apparatus for low latency data center network

A Universal Serial Bus (USB) switch includes a host port and a plurality of USB device ports. Each USB device port is configured to be coupled to a respective plurality of USB devices. A USB interface is coupled to the host port. A switch matrix is coupled between the USB interface and the plurality of USB device ports and is configured to switch the USB interface to one of the plurality of USB device ports. A switch controller is coupled to the plurality of USB interfaces, the host port, and the switch matrix. The switch controller is configured to control operation of the USB switch.

Universal serial bus network switch

A method implemented in a network element (NE) configured to operate in an information centric network (ICN) including receiving a test traffic flow as a copy of a traffic flow in a forwarding plane of the ICN during a defined interval, emulating the ICN at a state determined at an arrival time of the test traffic flow based on an active traffic flow in the ICN, measuring a first impact of a first candidate path for the test traffic flow in the emulated ICN to the active traffic flow, selecting the first candidate path based on the first impact measurement, and constructing a forwarding rule based on the state and the selection of the first candidate path.

Learning or emulation approach to traffic engineering in information-centric networks

A routing technique provides a routing table which assigns weights in the process of selecting a next hop at a router, while still using an equal cost multipath selection process at the router. The routing table is configured to cross reference an IP address prefix set to a number of next hops which can be all, or fewer than all, available next hops. This occurs in each row of the table for a different IP address prefix set. Subsets of the next hops are identified in each row in a manner which results in the next hops being selected according to specified weights. An estimate of traffic to the different IP address prefix set is also considered. The routing table can be configured based on announce and withdraw messages received from a link weight translator of a controller.

Weighted next hop selection at a router using an equal cost multipath process

An apparatus is configured to perform a method for congestion control in an Ethernet network. The method includes determining an egress queue congestion state for each of a plurality of egress queues. The method also includes determining an ingress queue congestion state for each of a plurality of ingress queues. The method further includes determining a flow control state for at least one of the ingress queues based on the determined egress queue congestion states and ingress queue congestion states. In addition, the method includes transmitting a flow control message to the at least one ingress queue based on the determined flow control state.

Method and apparatus for priority flow and congestion control in ethernet network

The disclosure relates to technology for load balancing link utilization of a networking device based on fractal analysis. In one embodiment, link utilization of switches, routers, etc. in a data center is balanced based on a fractal model of the link utilization. Techniques disclosed herein are proactive. For example, instead of reacting to link congestion, the technique predicts future link utilization based on fractal analysis. Then, packet flows (or flowlets) may be assigned to links based on the predicted future link utilization. Hence, congestion on links may be reduced or prevented.

Proactive load balancing based on fractal analysis

A method implemented in a network element (NE) configured to operate in an information centric network (ICN), comprising receiving a plurality of test traffic flows, wherein the test traffic flows are copies of a number of traffic flows received in a forwarding plane of the ICN during a defined interval; emulating the ICN at a state determined at an arrival time of the test traffic flows to the NE and defined based on a plurality of active traffic flows in the ICN; measuring a first impact of a plurality of first candidate paths for a first test traffic flow from the test traffic flows in the emulated ICN to the active traffic flows; selecting one of the first candidate paths based on the first impact measurement; measuring a second impact of a plurality of second candidate paths for a second test traffic flow from the test traffic flows in the emulated ICN to active traffic flows and the selected first candidate test path; selecting one of the second candidate paths based on the second impact measurement; and constructing a plurality of forwarding rules based on the state and the candidate paths selections.

Network element configured to operate in an information centric network

A method for reducing response times in an information-centric network includes receiving an indication from an ingress node of a content object entering a network, the content object associated with a new delivery flow through the network. An egress node in the network for the content object and a size of the content object are identified. A backlog and bandwidth for the new delivery flow is determined based in part on the size of the content object. Backlogs and bandwidths for existing delivery flows in the network are determined. A set of candidate paths in the network for the new delivery flow from the ingress node to the egress node is determined. For each candidate path, a total response time is estimated for completion of all delivery flows for each candidate path based on the backlog and bandwidth. The candidate path having the lowest total response time is selected for the new delivery flow.

Growing community of inventors

Milpitas, CA, United States of America

Serhat Nazim Avci