Facebook, Inc.

Meta Platforms, Inc.

Gayathrinath Nagarajan

Satyajeet Singh Ahuja

Nitin Kumar Goel

Petr V Lapukhov

Vinayak Dangui

Abishek Gopalan

Stephen Gregory Grubb

Yury Smirnov

Alexander Ilo Nikolaidis

Steve Politis

Srivatsan Balasubramanian

Ying Zhang

Soshant Bali

Herve Albert Pierre Fevrier

Varun Gupta

Hao Zhong

Sung Woo Choe

A system and method for network planning with certain guarantees is disclosed. The system receives data characterizing various aspects of a backbone network, such as the nodes of the backbone network, how the nodes are connected by network links, the maximum available capacities of the network assets, network costs, and network asset reliability information. The system also receives data characterizing the requirements of different data communications, or flows, within the backbone network. For example, the backbone network may need to provide a flow a minimum amount of bandwidth or throughput, and the flow may have a minimum required uptime or availability. Based on the network data and flow data, the system generates a network plan that describes how capacity should be provided by different components of the network in a manner that guarantees satisfying flow requirements while balancing other considerations, such as network costs.

Methods and systems for network planning with availability guarantees

The disclosed computer-implemented method may include (i) generating a data center constraint model by placing a constraint on a total amount of ingress or egress traffic a service expects from each respective data center of multiple data centers, (ii) filtering a set of traffic matrices that indicate points in the data center constraint model by comparing the set of traffic matrices against cut sets of a network topology that indicate network failures to create a tractable set of dominating traffic matrices, (iii) obtaining physical network resources to implement a cross-layer network upgrade architecture that satisfies the tractable set of dominating traffic matrices, and (iv) allocating the physical network resources across the multiple data centers according to the cross-layer network upgrade architecture such that a capacity level of the multiple data centers is increased while satisfying the data center constraint model. Various other methods, systems, and computer-readable media are also disclosed.

Systems and methods for configuring networks

A method and system for allocating network resources are described. The method includes receiving a plurality of forecasted network traffic patterns for a network. A representative subset of the plurality of forecasted network traffic patterns is selected based on an analysis of the plurality of forecasted network traffic patterns using a topology of the network. The selected representative subset of the plurality of forecasted network traffic patterns is used to determine a resource allocation for the network.

Network resource allocation using forecasted network traffic patterns selected utilizing network topology

Disclosed are fiber amplifiers with multiple pumping sources including multiple optical sources or an optical comb source with multiple spectral lines. A comb source may include generating a plurality of evenly spaced or nearly evenly spaced spectral lines. The optical comb source may pump a fiber by propagating optical energy at the multiple spectral lines through the fiber. The comb source may cause gain in the fiber at in a band of wavelengths different from the spectral lines of the comb source. A weak signal injected into the fiber that propagates in the fiber will experience optical gain in the fiber producing an amplified signal at the wavelength within a band of wavelengths different from the comb source wavelengths. When the bandwidth of the multiple bands of gain is wide, the amplifier may be referred to as an ultra-wideband amplifier.

Ultra-wideband Raman amplifier with comb source

Network planning with availability guarantees

A direction-switchable transponder of a high speed communications network, e.g., an fiber optic data communications network, is capable of dynamically reversing the data traffic flow of its various communications channels in response to a signal. The signal can specify a number of channels, a channel map, or a required bandwidth. The direction-switchable transponder can receive a signal relating to network bandwidth requirements; select, based on the received signal, one or more fiber optic channels for reversing direction of flow of network traffic; and dynamically and automatically reconfigure the selected fiber optic signal to reverse direction of flow of network traffic. By responding to asymmetric bandwidth requirements, the direction-switchable transponder uses high speed communications network lines more efficiently.

Direction-switchable transponders

Aspects of an optical communications network are described that include two or more optical fibers arranged to allow communication in the same direction. The optical network includes a first optical amplifier coupled to the first optical fiber, a second optical amplifier coupled to the second optical fiber, a first optical pump to provide optical power to the first optical fiber, and a second pump to provide optical power to both the first and the second optical fibers. By sharing the second pump between the first and the second optical fibers, a need to deploy additional pumps is alleviated. Scaling of the optical network to include additional optical fibers provides further cost savings by allowing more pumps to be shared among the multiple optical fibers.

Shared optical pumps among multiple fiber channels

The disclosed systems for multiple data center building optical communication may include (1) a first optical switching node of a first main point of entry (MPOE) of a first data center building that communicatively couples a first fiber pair of a first long-haul path to a computing system of the first building, (2) a second optical switching node of the first MPOE of the first building that communicatively couples a first fiber pair of a second long-haul path to the computing system of the first building, and (3) a third optical switching node of the first MPOE of the first building that communicatively couples the first and second optical switching nodes of the first MPOE of the first building to a second MPOE of the first building and a first MPOE of a second data center building. Various other systems and methods are also disclosed.

Systems and methods for multiple data center building optical communication

Submarine optical fiber communications are described. A submarine optical fiber communications system can be configured to transmit optical signals within a first band of the electromagnetic spectrum in a first direction, and to transmit optical signals within a second band of the electromagnetic spectrum in a second direction. The bands can be C band, L band, or both bands.

Submarine optical fiber communications architectures

A network topology is analyzed to identify shared risk link groups, the edge diversities of paths, and maximally diverse edges for paths. During operation of the network for conveying data packets between two end points, data flows are routed in the network by prioritizing the use of resources that do not belong to a shared risk group and are maximally diverse with other edges already being used. Various load balancing techniques can be used to minimize the risk of serious disruption in the event an underlying resource of a shared risk link group goes down.

Growing community of inventors

Sunnyvale, CA, United States of America

Gayathrinath Nagarajan