公开(公告)号：US11601344B1

公开(公告)日：2023-03-07

申请号：US17564732

申请日：2021-12-29

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Zosim Kanevsky ,  Wen-Jui Li

IPC: H04L43/0817 ,  H04L41/50 ,  H04L41/5009 ,  H04L41/0604

Abstract: A cloud gateway outage risk detector can receive, by an event listener module, user session data associated with a plurality of user sessions over a cloud gateway. The event listener module can store the data in a database. A run-time collection module can obtain at least a portion of the data. The run-time collection module can provide the portion of the data to a run-time risk criteria evaluation module that can determine, based upon the portion of the data, a run-time outage risk criteria for the cloud gateway. A baseline risk criteria evaluation module can obtain historical data from the database. The baseline risk criteria evaluation module can determine, based upon the data, a baseline outage risk criteria for the cloud gateway. The run-time risk criteria evaluation module can determine whether the run-time outage risk criteria meets or exceeds an outage risk threshold.

公开(公告)号：US10673877B2

公开(公告)日：2020-06-02

申请号：US16505455

申请日：2019-07-08

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Wai Sum Lai ,  Andrew Egan ,  Wen-Jui Li

IPC: G06F21/31 ,  H04L29/06 ,  H04L12/26

Abstract: A method and an apparatus for detecting a port scan in a network are disclosed. For example, the method extracts statistics from a message, detects the port scan for a source internet protocol address, determines whether a port scan record exists for the source internet protocol address, creates a port scan record for the source internet protocol address that is extracted when the port scan record does not exist, determines an elapsed time when the port scan record does exist, wherein the elapsed time is determined as a difference between the time stamp that is extracted and a recorded time stamp, sets the recorded time stamp to be the extracted time stamp when the elapsed time is less than an intra-scan time, and determines the port scan has ended for the source internet protocol address when the elapsed time is not less than the intra-scan time.

公开(公告)号：US20180316594A1

公开(公告)日：2018-11-01

申请号：US15499452

申请日：2017-04-27

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Tsong-Ho Wu ,  Wen-Jui Li ,  Wai Lai

IPC: H04L12/751 ,  H04L12/931 ,  H04L12/24

CPC classification number: H04L45/02 ,  H04L41/12 ,  H04L45/64

Abstract: An IP network topology update system may update IP network topology in near real-time and on-demand with minimum overheads. It identifies likely impact area (e.g., layer 2 or layer 3), objects (e.g., link or node such a device), and timing (e.g., what topology objects located where or when the topology update process should be performed) in the IP Layer 3 network and its underlying SDN Layer 2 network under virtualized networking infrastructure as candidates of impacts for topology update

公开(公告)号：US20170295082A1

公开(公告)日：2017-10-12

申请号：US15092681

申请日：2016-04-07

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Tsong-Ho Wu ,  Wen-Jui Li ,  Hasan Erkan

IPC: H04L12/26 ,  G06F9/455 ,  H04L29/08

CPC classification number: H04L43/14 ,  G06F8/38 ,  G06F9/45558 ,  G06F2009/4557 ,  G06F2009/45591 ,  G06F2009/45595 ,  H04L41/0816 ,  H04L41/5009 ,  H04L43/08 ,  H04L43/16 ,  H04L67/10

Abstract: Concepts and technologies disclosed herein are directed to an auto-scaling software-defined monitoring (“SDM”) platform for software-defined networking (“SDN”) service assurance. According to one aspect of the concepts and technologies disclosed herein, an SDM controller can monitor event data associated with a network even that occurred within a virtualized IP SDN network that is monitored by a virtualized SDM resources platform. The SDM controller can measure, based upon the event data, a quality of service (“QoS”) performance metric associated with the virtualized SDM resource platform. The SDN controller can determine, based upon the QoS performance metric, whether an auto-scaling operation is to be performed. The auto-scaling operation can include reconfiguring the virtualized SDM resources platform by adding virtual machine capacity for supporting event management tasks either by instantiating a new virtual machine or by migrating an existing virtual machine to a new hardware host.

公开(公告)号：US20190356535A1

公开(公告)日：2019-11-21

申请号：US15981305

申请日：2018-05-16

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Wen-Jui Li ,  Tsong-Ho Wu ,  Wai Sum Lai ,  Zosim Kanevsky ,  Jun-Min Liu

IPC: H04L12/24 ,  H04L12/26

Abstract: Concepts and technologies directed to network fault originator identification for virtual network infrastructure are disclosed herein. Embodiments can include a control system that is communicatively coupled with network infrastructure. The control system can include a processor and memory that, upon execution, causes the control system to perform operations. The operations can include determining, based on a source ticket, a network fault condition associated with the network infrastructure. The operations can further include identifying, from the source ticket, a trap set and an alarm set that are associated with origination of the network fault condition. The operations can include the control system collecting network event data from the network infrastructure prior to a polling time of a fault reporting schedule; determining that a qualified source ticket should be created; and generating the qualified source ticket based on the network event data.

公开(公告)号：US20190334937A1

公开(公告)日：2019-10-31

申请号：US16505455

申请日：2019-07-08

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Wai Sum Lai ,  Andrew Egan ,  Wen-Jui Li

IPC: H04L29/06 ,  H04L12/26

Abstract: A method and an apparatus for detecting a port scan in a network are disclosed. For example, the method extracts statistics from a message, detects the port scan for a source internet protocol address, determines whether a port scan record exists for the source internet protocol address, creates a port scan record for the source internet protocol address that is extracted when the port scan record does not exist, determines an elapsed time when the port scan record does exist, wherein the elapsed time is determined as a difference between the time stamp that is extracted and a recorded time stamp, sets the recorded time stamp to be the extracted time stamp when the elapsed time is less than an intra-scan time, and determines the port scan has ended for the source internet protocol address when the elapsed time is not less than the intra-scan time.

公开(公告)号：US20190102717A1

公开(公告)日：2019-04-04

申请号：US15720396

申请日：2017-09-29

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Tsong-Ho Wu ,  Wen-Jui Li ,  Jun-Min Liu

IPC: G06Q10/06 ,  H04L12/24

Abstract: Methods and systems associated with a microservice based predictive service level agreement (SLA) impact analytics system that may run on standardized container based virtual computing platform to enable capacity auto-scaling for on-demand, near-real-time resource allocation automatically supporting user data packet forwarding when SLA is potentially impacted to ensure SLA compliance.

公开(公告)号：US20190081879A1

公开(公告)日：2019-03-14

申请号：US16190470

申请日：2018-11-14

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Tsong-Ho Wu ,  Wen-Jui Li ,  Hasan Erkan

IPC: H04L12/26 ,  H04L12/24 ,  G06F9/455 ,  G06F8/38 ,  H04L29/08

CPC classification number: H04L43/14 ,  G06F8/38 ,  G06F9/45558 ,  G06F2009/4557 ,  G06F2009/45591 ,  G06F2009/45595 ,  H04L41/0816 ,  H04L41/5009 ,  H04L43/08 ,  H04L43/16 ,  H04L67/10

Abstract: Concepts and technologies disclosed herein are directed to an auto-scaling software-defined monitoring (“SDM”) platform for software-defined networking (“SDN”) service assurance. According to one aspect of the concepts and technologies disclosed herein, an SDM controller can monitor event data associated with a network event that occurred within a virtualized IP SDN network that is monitored by a virtualized SDM resources platform. The SDM controller can measure, based upon the event data, a quality of service (“QoS”) performance metric associated with the virtualized SDM resource platform. The SDN controller can determine, based upon the QoS performance metric, whether an auto-scaling operation is to be performed. The auto-scaling operation can include reconfiguring the virtualized SDM resources platform by adding virtual machine capacity for supporting event management tasks either by instantiating a new virtual machine or by migrating an existing virtual machine to a new hardware host.

公开(公告)号：US20190081869A1

公开(公告)日：2019-03-14

申请号：US16186903

申请日：2018-11-12

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Tsong-Ho Wu ,  Wen-Jui Li

IPC: H04L12/24 ,  H04L12/26 ,  H04L12/715 ,  H04L12/923 ,  H04L29/08 ,  H04L29/06

CPC classification number: H04L41/5067 ,  H04L41/0631 ,  H04L41/0896 ,  H04L41/12 ,  H04L41/5009 ,  H04L41/5025 ,  H04L41/509 ,  H04L43/0817 ,  H04L43/0876 ,  H04L43/0894 ,  H04L45/64 ,  H04L47/762 ,  H04L65/80 ,  H04L67/327

Abstract: Concepts and technologies disclosed herein are directed to context-aware virtualized control decision support system (“DSS”) for providing quality of experience (“QoE”) assurance for Internet protocol (“IP”) streaming video services. A QoE assurance DSS can monitor QoE event and context data to be utilized for QoE assurance analytics, measure QoE performance, perform QoE assurance analytics, and determine whether the QoE assurance analytics indicate that the QoE has been degraded, and if so, construct a fault correlation information model to be utilized for root cause analysis to determine a root cause of the QoE being degraded. The QoE assurance DSS also can determine, based upon the fault correlation information model, whether the root cause of the QoE being degraded is due to a capacity reduction, and if so, the QoE assurance DSS can identify a new network resource for capacity reallocation to accommodate a virtual machine migration.

公开(公告)号：US20170024234A1

公开(公告)日：2017-01-26

申请号：US14806906

申请日：2015-07-23

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Tsong-Ho Wu ,  Wen-Jui Li

IPC: G06F9/455 ,  G06F9/46

CPC classification number: G06F9/45558 ,  G06F9/45533 ,  G06F9/466 ,  G06F2009/45595

Abstract: Concepts and technologies disclosed herein are directed to data-driven feedback control system for an acceptable level of real-time application transaction completion rate in virtualized networks, while maximizing virtualized server utilization. According to one aspect disclosed herein, a network virtualization platform (“NVP”) includes a plurality of hardware resources, a virtual machine (“VM”), and a virtual machine monitor (“VMM”). The VMM can track an execution state of each of a plurality of applications associated with the VM. The VMM can measure a real-time application transaction completion rate of the VM. The VMM can determine whether a trigger condition exists for priority scheduling of real-time applications based upon the real-time application transaction completion rate and a pre-set threshold value. The VMM can, in response to determining that the trigger condition exists, apply a priority control schedule to instruct the VM to perform priority processing of a real-time application over a non-real-time application.

Abstract translation: 本文中公开的概念和技术涉及数据驱动的反馈控制系统，以实现虚拟化网络中实时应用事务完成率的可接受水平，同时最大化虚拟化服务器利用率。 根据本文公开的一个方面，网络虚拟化平台（“NVP”）包括多个硬件资源，虚拟机（“VM”）和虚拟机监视器（“VMM”）。 VMM可以跟踪与VM相关联的多个应用中的每一个的执行状态。 VMM可以测量VM的实时应用程序事务完成率。 VMM可以基于实时应用事务完成率和预设阈值来确定是否存在针对实时应用的优先级调度的触发条件。 响应于确定触发条件存在，VMM可以应用优先级控制调度来指示VM通过非实时应用执行实时应用的优先处理。