公开(公告)号：US20250055906A1

公开(公告)日：2025-02-13

申请号：US18799614

申请日：2024-08-09

Applicant: Netflix, Inc.

Inventor： Arthur Gonigberg ,  Argha Chattopadhyay

IPC: H04L67/1008 ,  H04L67/1014

Abstract: The disclosed computer-implemented method may include receiving, by a network gateway service, a service request from a client device. The method may also include assigning, by the network gateway service, the service request to an event loop. Additionally, the method may include determining, by the network gateway service, a destination of the event loop based on one or more resources of one or more nodes in a ring of nodes. Furthermore, the method may include creating, by the network gateway service, a connection pool for the event loop to select a node subset of the ring of nodes based on a mapping of event loops to node subsets. Finally, the method may include performing, by the network gateway service, load balancing of the connection pool to execute the service request. Various other methods, systems, and computer-readable media are also disclosed.

公开(公告)号：US11929869B2

公开(公告)日：2024-03-12

申请号：US16410867

申请日：2019-05-13

Applicant: NETFLIX, INC.

Inventor： Arthur Gonigberg ,  Michael Lloyd Cohen ,  Neeraj Joshi ,  Cody Mitchell Rioux

IPC: H04L41/0681 ,  H04L41/0604 ,  H04L41/0631 ,  H04L43/04 ,  H04L43/0817 ,  H04L43/0823 ,  H04L43/16

CPC classification number: H04L41/0681 ,  H04L41/0609 ,  H04L41/0631 ,  H04L43/04 ,  H04L43/0817 ,  H04L43/0823 ,  H04L43/16

Abstract: Various embodiments of the disclosure disclosed herein provide techniques for detecting anomalies across one or more components within a distributed computing system, according to various embodiments of the present disclosure. An anomaly detection system retrieves event data associated with a real-time stream of events generated by one or more components within a distributed computing system. The anomaly detection system computes a failure metric based on the event data. The anomaly detection system determines that the failure metric exceeds a dynamically adjustable trigger condition. The anomaly detection system generates an alert associated with the failure metric.