公开(公告)号：US12284110B1

公开(公告)日：2025-04-22

申请号：US17853531

申请日：2022-06-29

Applicant: Amazon Technologies, Inc.

Inventor： Joan Barcelo Llado ,  Eric Charles Briffa ,  Bobby Brown ,  Kenan Yang ,  Seyed Arash Sadrieh ,  Tahir Azim ,  Julien Ridoux ,  Harvo Reyzell Jones ,  Thomas Bradley Scholl ,  Ali Khayam ,  Robert Louis Mosher ,  Michael W. Palladino ,  Rajat Chaudhary ,  Nikhil Ashok Arjunagi ,  Fiachra John Groarke

IPC: H04L45/12 ,  H04L43/087 ,  H04L43/0888 ,  H04L45/121 ,  H04L47/125

Abstract: Technologies are disclosed for performing context-aware routing of traffic based on an application of a series of filters and/or selectors to determine an optimal group of interfaces for servicing traffic relating to a prefix. A traffic management control plane processes input data corresponding to aggregated telemetry data for a network by determining candidate groups of interfaces to service a network prefix and applies tunable filters and/or selectors to the candidate groups to determine a targeted group to use for servicing traffic corresponding to the prefix. The determined targeted group is mapped to the prefix in a mapping result, which is then provided to an actuator interface to control network devices to inject the new mapping into the network to control traffic according to the mapping.

公开(公告)号：US11343207B1

公开(公告)日：2022-05-24

申请号：US16871817

申请日：2020-05-11

Applicant: Amazon Technologies, Inc.

Inventor： John William Evans ,  Seyed Arash Sadrieh ,  Alber Edmon Abdelshahid Moines ,  Colin John Whittaker

IPC: H04L49/90 ,  H04L47/50 ,  H04L41/14 ,  H04L47/32 ,  H04L41/0816 ,  H04L47/30

Abstract: An optimization agent optimizes configuration parameters of a shared memory buffer for ports of a network device (e.g., router). The optimization agent located on the network device (or remotely) takes an input state from the network device's switching logic, such as the packets transmitted or dropped per port, and determines port buffer settings to optimize the settings to achieve a goal. The buffer settings can include minimum and maximum buffer limits per port and the goal can be to maximize port utilization or minimize port loss. The optimization agent can then take an action by applying the buffer configurations to the switching logic and repeat the process iteratively until the goal is reached.

公开(公告)号：US11294756B1

公开(公告)日：2022-04-05

申请号：US16575986

申请日：2019-09-19

Applicant: Amazon Technologies, Inc.

Inventor： Seyed Arash Sadrieh ,  Joan Enric Barcelo Llado

IPC: G06F11/00 ,  G06F11/07 ,  G06N3/04

Abstract: Anomaly detection for one or more streams of time-series data can use an encoder/decoder pair, such as in a variational autoencoder (VAE) in combination with an aggregator or classifier, such as a random isolation forest (RIF). A particular application relates to detecting anomalies in network updates in a large number of network devices that can transmit the updates to a collector for analysis. The encoder/decoder pair can include a neural network with long short-term memory cells or similar type cells. Using the combination, a single anomaly score can be produced from multiple streams of the time-series data.