Amazon Technologies, Inc.

Stephen Callaghan

Leonard Thomas Tracy

Justin Oliver Pietsch

Mark Noel Kelly

Alan Michael Judge

Frederick David Sinn

Edward Crabbe

Amit Sahoo

Spencer Giacalone

Nafea Bshara

Diwakar Gupta

Thomas A Volpe

Mark Bradley Davis

Kiran Kumar Edara

Ishwardutt Parulkar

Thomas Bradley Scholl

Joseph Elmar Magerramov

Kaixiang Hu

Shane Ashley Hall

Igor A Kostic

Colin John Whittaker

Ethan Joseph Torretta

John David Backes

Matthew Dean Rehder

Viktor Heorhiadi

Ashi Ramachandran Sudhakumari

Minsung Kim

Nikolaos Giannarakis

Giacomo Bernardi

Owais Bin Zuber

Maciej Rzehak

Japson Jeyasekaran

Pier Maria Mattioli

Arti Bhatt

Pavel Kiselev

Network management systems and associated methods are described for providing intent-driven management of networks using one or more compilers. An intent identifying an expectation for behavior in a network and entity associations for the intent may be received at a compiler. The compiler may determine one or more artifacts for the respective entities based on the intent. The compiler may output instructions based on the artifacts for the consumers. The instructions may be in a language and/or format corresponding to the consumers.

Intent-driven network management

To achieve impact-less re-routing of traffic towards non best-paths in a network, two locator routes are originated for each network fabric: a primary locator and a secondary locator. In a steady state, the forwarding state on each router is created by recursion of endpoints identifiers through their originating location's primary locator-route. Secondary locator routes are not used in the steady state. At a desired point, a steering route is injected into the network device to redirect routing from the primary locator route to the secondary locator route. The secondary locator route has a different resultant port from the primary locator route to redirect incoming packets on a different network path.

Shifting of network traffic using converged primary and secondary paths in a network device

Disclosed are various embodiments that provide highly available data-processing network functions for radio-based networks. In one embodiment, a tunnel host consistently routes network traffic associated with a range of network addresses in a radio-based network to a first instance of a data-processing network function instead of a second instance of the data-processing network function. A problem with the first instance of the data-processing network function is then detected. Additional network traffic associated with the range of network addresses is redirected from the first instance of the data-processing network function to the second instance of the data-processing network function.

Highly available data-processing network functions for radio-based networks

Technologies for adaptively classifying network packets for quality of service (QoS) are provided herein. Adaptive classifiers track network packets transferred in data flows and network sessions, and apply various QoS classifications to the packets. An adaptive classifier can apply different QoS markings to packets in a data flow based on data flow metadata, such as the age of the data flow and/or the amount of data transmitted in the data flow. An adaptive classifier can assign various priority levels to network packets during different phases of a network session. Routers that receive the packets can prioritize transmission of the packets based on the applied QoS markings or assigned priority levels.

Adaptive quality of service marking

Remediation of network devices that are failing is accomplished using a Shared Failure Domain (SFD) database that provides neighboring device/link information to remediation tools. SFD refers to a group of objects (links/devices) that share a same failure model. A state change of one or multiple of the objects results in a corresponding action on other devices linked together through the SFD. Moreover, the SFD data is available in a central repository and software tools consult the central repository for failure domain data before taking remedial actions. SFD data is generated using configuration generation and device state. Software tools lookup SFD data during operational events (device/link down) and take appropriate actions on the neighboring devices.

Configuring network devices using a shared failure domain

A network switch and system for detecting the capacity of available paths in a network and for modifying traffic distribution at each stage of the network, such that traffic is proportionally balanced across the unequal capacity paths. A centralized controller manipulates path weights on each switch such that a load is distributed proportionately to the capacity available to ensure that all available capacity is evenly utilized. A central view of the network is used to determine capacity information, calculate a minimal change set to optimize the traffic flows, and modify the existing multipath group objects. A centralized application can use the capacity information of each switch to build a capacity model of the network. Once the full model has been built up by the application, programming of the specific decisions is done via the controller through an API in communication with each network device's local agent.

Network switches using centralized improvement of routing based on link capacity

Client device prefixes are distributed to other network devices (e.g., routers) in the network inside a special container attribute of an update message. The container attribute is attached as an optional-transitive attribute to a location prefix announcement. A location prefix identifies the Autonomous System (AS) that originates the client prefixes. All of the ID prefixes packed in the container use the location prefix as the recursive next-hop towards them. As a result, all convergence in the network occurs for a small number of location prefixes, while at the same time, the location/ID mapping is being distributed using the Location best-paths. In some embodiments, the container attribute can utilize BGP update packets, or optimized NLRI/attribute packaging. Compression of the ID prefixes can also be used within the messages.

Distribution of prefixes in a network

The following description is directed to managing a hierarchical network including one or more network zones. In one example, a method of managing a hierarchical network includes collecting network state from respective devices of a network zone; using the collected network state to build an aggregated topology of the network zone; and transmitting the aggregated topology of the network zone to a traffic engineering service.

Managing a hierarchical network

Network packet tracing may be implemented on packet processors or other devices that perform packet processing. As network packets are received, a determination may be made as to whether tracing is enabled for the network packets. For those network packets with tracing enabled, trace information may be generated and the network packets modified to include the trace information such that forwarding decisions for the network packets ignore the trace information. Trace information indicate a packet processor as a location in a route traversed by the network packets and may include ingress and egress timestamps. Forwarding decisions may then be made and the network packets sent according to the forwarding decisions. Tracing may be enabled or disabled by packet processors for individual network packets. Trace information may also be truncated at a packet processor.

Network packet tracing

Network links are exercised by transmitting network frames across the links at a threshold level of network traffic by filling unutilized bandwidth with test frames. By increasing network traffic across the links, errors are increased and easier to detect. The test frames can be given lower priority than customer traffic so as not to impact the test traffic. The test frames can be designed such that they are dropped upon receipt by another network switch.

Network link testing

Methods, systems, and devices are described for transmitting and storing routing path information and routing topology information using a single protocol. In particular embodiments, routing path information for an exterior gateway protocol ('EGP') network such as a border gateway protocol ('BGP') network is transmitted throughout an interior gateway protocol (&#x201c;IGP&#x201d;) network using IGP messages such as open-shortest-path-first (&#x201c;OSPF&#x201d;) or intermediate state to intermediates state (&#x201c;IS-IS&#x201d;) messages. The IGP networks may transmit and store the BGP information using type length values (TLVs). As a result, network areas running an IGP may maintain BGP information throughout the network area without the overhead of an iBGP mesh and related message-passing.

Transmission and storage of routing path information and routing topology information using a single routing protocol

In some embodiments, a link monitoring system ('LMS') is configured to obtain link status metrics and packet transmission metrics for network links forming a network channel between a first network device and a second network device. The LMS can obtain link status metrics from the first network device and the second network device and can also initiate the generation of packets on the network links to obtain the packet transmission metrics. Based on the status metrics and packet transmission metrics, the LMS can determine if the network links are malfunctioning.

Network link monitoring and testing

A method for adjusting capacity in a multi-stage routing network includes monitoring a number of available connections between a router in a first stage of a multi-stage router network and one or more routers in a second stage of the multi-stage router network. Each of the stages of the multi-stage router network may include a plurality of routers. The method may also include detecting that the number of available connections falls below a threshold number. A notification can be sent to one or more routers in a third stage of the multi-stage router network that the router in the first stage is deprioritized. The one or more routers in the third stage can be operated so that communications to the first stage are routed to one or more other routers in the first stage.

Adjusting communication capacity in a multi-stage routing network

Techniques and solutions for distributing routing updates using timing information are described. For example, inter-device messaging protocols can support timing information (e.g., routing protocols be extended to support timing information). The timing information can indicate a time at which the routing update is to be put into effect where the time decreases the further away a given routing device is from the source of the routing update. For example, each routing device can determine a decayed update reference time when the given routing device will put the routing update into effect according to a decay mode. Routing updates can be managed between routing devices that support timing information and routing devices that do not support timing information.

Growing community of inventors

Seattle, WA, United States of America

Stephen Callaghan