Google Inc.

Xiaoxue Zhao

Hong Liu

Amin Vahdat

Bikash Koley

Paul Germano

Ryohei Urata

Vijay Vusirikala

Emilie Jeanne Anne Danna

Philip Michael Wells

Edward Crabbe

Chris Johnson

Christoph Albrecht

Cedric Fung Lam

Satyajeet Singh Ahuja

Leon Poutievski

Vinayak Dangui

Wenjie Jiang

Ajay Kumar Bangla

Alireza Ghaffarkhah

Benjamin Preskill

The present disclosure describes system and methods for network planning. The systems and methods can incorporate network traffic demands, availability requirements, latency, physical infrastructure and networking device capability, and detailed cost structures to calculate a network design with minimum or reduced cost compared to conventional methods. In some implementations, the method include providing an initial, deterministic set of failures, and then successively performing a network optimization and a network availability simulation to determine which failures most impact the performance of the network model. The high impact failures can then be provided back into the system, which generates an improved network design while still maintaining minimum cost.

Network stochastic cross-layer optimization for meeting traffic flow availability target at minimum cost

The present disclosure provides a probabilistic framework that can calculate the probability of fulfilling demands for a given set of traffic flows. In some implementations, the probability of fulfilling demands can be based on the probability of infrastructure component failures, shared risk link groups derived from a cross-layer network topology, and traffic engineering (TE) considerations. The consideration of the cross-layer network topology enables the systems and methods described herein to account for the relationship between the physical and logical topologies.

Parallelized network traffic flow availability simulation using stochastic process and traffic engineering algorithms

According to at least one aspect, a network system includes a wavelength selective switch (WSS) mesh network, multiple dense wavelength division multiplexing (DWDM) multiplexers/de-multiplexers, and a controller. The WSS mesh network includes a plurality of WSS components. Each WSS component includes a common port and multiple switching ports. The plurality of WSS components are coupled to each other through corresponding switching ports. Each DWDM multiplexer/de-multiplexer is coupled to one of multiple sets of servers via a respective top tier switch and respective aggregation nodes and coupled to a common port of a WSS component of the WSS mesh network. The controller is configured to obtain indications of communication bandwidth demand for communications between the multiple sets of servers, determine a configuration for the WSS mesh network based on the obtained indications of communication bandwidth demand, and cause the WSS components and the top tier switches to be configured according to the determined configuration.

Dynamic data center network with a mesh of wavelength selective switches

Methods, mediums and systems described herein determine real-time in-service OSNR measurement without disrupting or turning off one or more channels of the network. An OSNR monitor described herein may be integrated with existing optical line systems. The OSNR measurements performed by the OSNR monitors are independent of the modulation format and thus, may work with all phase formats, amplitude format or a combination thereof. The real-time in-service OSNR data may be used to perform global network optimization to determine the optimal routing and data rate in the optical network. The OSNR data may be used to establish protection and restoration paths for network resiliency and to maximize data throughput.

In-service optical signal to noise ratio monitoring in an optical network

The specification describes an apparatus including a plurality of communication ports, each communication port coupled to a corresponding networking element, and a plurality of optical transceivers, each optical transceiver coupled to a corresponding communication port and including an optical receiver and an optical transmitter, wherein the optical transmitter can transmit an optical data signal having a carrier wavelength different than any other of the plurality of optical transmitters. A passive optical multiplexer/demultiplexer module is coupled to the transceivers and includes an optical multiplexer including a wavelength division multiplexing (WDM) output and a plurality of inputs, each input coupled to one of the plurality of transmitters, and an optical demultiplexer including a WDM input and a plurality of outputs, each output coupled to one of the plurality of receivers. Placing the passive optical multiplexer and demultiplexer external to the optical transceivers creates an aggregation/de-aggregation point with extremely low fault probability.

Aggregation and de-agreggation of bandwidth within data centers using passive optical elements

Embodiments provide a methodology for designing a large-scale non-blocking OCS using a multi-stage folded CLOS switch architecture for use in datacenter networks and fiber-rich backbone network POPs. One aspect employs a folded CLOS architecture because of its ease of implementation, enabling the topology to scale arbitrarily with increasing number of stages. The fraction of ports allocated for internal switch wiring (overhead) also increases with the number of stages. Design decisions are made to carefully optimize the insertion loss per module, number of ports per module, number of stages and the total scale required. Other embodiments include folded CLOS switch architectures having at least two stages. In one example, power monitoring may be included only on the leaf switches.

Implementation of a large-scale multi-stage non-blocking optical circuit switch

A system and method of providing a dynamic optical network topology according to topology determinations made by a network control is disclosed. The system and method includes optical ports on an optical circuit switch system operably connected to a plurality of server groups, and optical ports on the optical circuit switch system operably connected to a plurality of packet processing nodes. The system and method also includes at least one memory and at least one processor to execute network control software to receive input comprising a bandwidth request, determine an output comprising a preferred optical link topology for the optical circuit switch system based on the received input, convert the optical link topology for the optical circuit switch system into optical circuit switch port mapping, and send the optical circuit switch port mapping to the optical circuit switch system and to the packet processing nodes.

Dynamic data center network with optical circuit switch

A packet switch/router including a first stage switch fabric receiving an electrical signal, a mid-stage buffer receiving and storing the electrical signal from the first stage switch fabric, and a second stage switch fabric receiving the electrical signal from the mid-stage buffer. Each switch fabric includes N layers of N&#xd7;N arrayed waveguide gratings (AWGs), and each AWG has ingress ports and egress ports. A wavelength tunable device, such as a tunable laser, communicates with a source ingress port of an AWG and converts the received electrical signal to an optical signal having a wavelength selected for routing a packet from the source ingress port to a target egress port of the arrayed waveguide grating. A photoreceiver, such as a burst-mode photoreceiver, receives the propagated optical signal from the target egress port and converts the optical signal to the electrical signal.

Optoelectronic packet switch/routers

The disclosure describes implementations of an apparatus including a plurality of racks, wherein each rack houses a plurality of networking devices and each networking device includes a communication port. An optical circuit switch can be coupled to each of the plurality of communication ports in one or more of the plurality of racks, and a plurality of top-of-rack (TOR) switches can be coupled to the optical circuit switch. Other implementations are disclosed and claimed.

Flexible non-modular data center with reconfigurable extended-reach optical network fabric

Systems and methods for optimizing port usage in an optical circuit switch are disclosed herein. A plurality of optical circulators can be coupled to the plurality of input and output ports of an optical circuit switch. An optical circulator coupled to an input port and an optical circulator coupled to an output port can form a bidirectional pair capable of communicating with any other bidirectional pair of the optical circuit switch. In this regard, the number of bidirectional ports of an optical circuit switch can be increased allowing additional bandwidth to flow through the optical circuit switch. Increasing the efficiency and capabilities of optical circuit switches within a large scale communications network can offer increased functionality and performance with little trade off.

Growing community of inventors

Fremont, CA, United States of America

Xiaoxue Zhao