One embodiment of the present invention provides a switch. The switch includes a tunnel management module, a packet processor, and a forwarding module. The tunnel management module operates the switch as a tunnel gateway capable of terminating an overlay tunnel. During operation, the packet processor, which is coupled to the tunnel management module, identifies in a data packet a virtual Internet Protocol (IP) address associated with a virtual tunnel gateway. This virtual tunnel gateway is associated with the switch and the data packet is associated with the overlay tunnel. The forwarding module determines an output port for an inner packet in the data packet based on a destination address of the inner packet.

Overlay tunnel in a fabric switch

One embodiment of the present invention provides a switch system. The switch includes one or more ports on the switch configured to transmit packets encapsulated based on a first protocol. The switch further includes a control mechanism. During operation, the control mechanism forms a logical switch based on a second protocol, receives an automatically assigned identifier for the logical switch without requiring manual configuration of the identifier, and joins a virtual cluster switch.

Virtual cluster switching

One embodiment of the present invention provides a system for facilitating flow definition management in a switch. During operation, the system identifies a generic flow definition which specifies a flow that is not specific to any input port of a switch. The system further stores in a flow lookup data structure one or more port-specific flow rules based on the generic flow definition, wherein each port-specific flow rule corresponds to a respective port capable of processing data flows.

System and method for flow management in software-defined networks

Methods and techniques for flooding packets on a per-virtual-network basis are described. Some embodiments provide a method (e.g., a switch) which determines an internal virtual network identifier based on one or more fields in a packet's header. Next, the method performs a forwarding lookup operation based on the internal virtual network identifier. If the forwarding lookup operation succeeds, the method can process and forward the packet accordingly. However, if the forwarding lookup operation fails, the method can determine a set of egress ports based on the internal virtual network identifier. Next, for each egress port in the set of egress ports, the method can flood the packet if a virtual network identifier in the packet's header is associated with the egress port. Flooding packets on a per-virtual-network basis can substantially reduce the amount of resources required to flood the packet when a forwarding lookup operation fails.

Flooding packets on a per-virtual-network basis

One embodiment of the present invention provides a switch. The switch includes a link aggregation module and a link management module. The link aggregation module establishes a virtual link aggregation group comprising a plurality of links coupled to the switch and one or more other switches. The plurality of links includes a first and a second sets of links coupling a first and a second end devices, respectively. The link management module determines a current mode which indicates which of the sets of links is currently active, and operates the first and the second sets of links as active and standby links, respectively, for the virtual link aggregation group based on the current mode and a port role for a port participating in the virtual link aggregation group. The port role indicates whether the port is coupled to an active link or a backup link.

Redundent virtual link aggregation group

In one aspect, a system includes one or more electrical switches to transfer data in a data network; one or more optical switching groups coupled to each electrical switch, each switching group having one or more server racks, each server rack coupled to a top of rack (TOR) switch and an optical transceiver coupled to the TOR switch; and an optical switching unit (OSU) coupled to the one or more optical switching groups.

Hybrid optical/electrical switching system for data center networks

Systems and methods are disclosed for a method to communicate over an optical network by using hop-by-hop routing over an optical network; and dynamically constructing a network topology.

Optical switching network

Network tomography has been proposed to ascertain internal network performances from end-to-end measurements. In this work, we present \emph{priority probing}, an optimal probing scheme for unicast network delay tomography that is proven to provide the most accurate estimation. We first demonstrate that the Fisher information matrix in unicast network delay tomography can be decomposed into an additive form where each term can be obtained numerically. This establishes the space over which we can design the optimal probing scheme. Then, we formulate the optimal probing problem into a semi-definite programming (SDP) problem. High computation complexity constrains the SDP solution to only small scale scenarios. In response, we propose a greedy algorithm that approximates the optimal solution. Evaluations through simulation demonstrate that priority probing effectively increases estimation accuracy with a fixed number of probes.

Optimal Probing for Unicast Network Delay Tomography

An optical switching system is described. The system includes a plurality of interconnected wavelength selective switching units. Each of the wavelength selective switching units is associated with one or more server racks. The interconnected wavelength selective switching units are arranged into a fixed structure high-dimensional interconnect architecture comprising a plurality of fixed and structured optical links. The optical links are arranged in a k-ary n-cube, ring, mesh, torus, direct binary n-cube, indirect binary n-cube, Omega network or hypercube architecture.

Method and apparatus for implementing a multi-dimensional optical circuit switching fabric

A method and a system are disclosed for determining application dependency paths in a data center. The method and the system captures application traffic volume data on the servers with switches and monitoring agents; generates an application traffic matrix of all the components of the applications based on the application traffic volume data; estimates the number of the applications in the data center from the traffic matrix with a Rank Estimation via Singular Value Decomposition or Power Factorization Residue Errors process; and decomposes the traffic matrix into a first matrix and a second matrix with a non-negative matrix factorization process using the estimated number of applications. The first matrix represents a set of the components belonging to each of the applications and the second matrix represents the amount of traffic generated by each application over time. Any noise in the first and second matrices is removed with a concurrent volumes ratios based correlation process.

Yueping Zhang

Papers

Hybrid optical/electrical switching system for data center networks

Optical switching network

Optimal Probing for Unicast Network Delay Tomography

Method and apparatus for implementing a multi-dimensional optical circuit switching fabric

System and method for determining application dependency paths in a data center