公开(公告)号：US20140281739A1

公开(公告)日：2014-09-18

申请号：US13826942

申请日：2013-03-14

Applicant: NETFLIX, INC.

Inventor： Philip Simon Tuffs ,  Roy Rapoport ,  Ariel Tseitlin

IPC: G06F11/34

CPC classification number: G06F11/3452 ,  G06F11/0709 ,  G06F11/079 ,  G06F11/3409 ,  G06F11/3495

Abstract: Techniques are described for identifying a root cause of a pattern of performance data in a system including a plurality of services. Embodiments provide dependency information for each of the plurality of services, where at least one of the plurality of services is dependent upon a first one of the plurality of services. Each of the plurality of services is monitored to collect performance data for the respective service. Embodiments further analyze the performance data to identify a cluster of services that each follow a pattern of performance data. The first one of the services in the cluster of services is determined to be a root cause of the pattern of performance data, based on the determined dependency information for each of the plurality of services.

Abstract translation: 描述了用于识别包括多个服务的系统中的性能数据模式的根本原因的技术。 实施例为多个服务中的每一个提供依赖信息，其中多个服务中的至少一个服务依赖于多个服务中的第一个服务。 监视多个服务中的每一个以收集相应服务的性能数据。 实施例进一步分析性能数据，以识别每个遵循性能数据模式的服务集群。 基于针对多个服务中的每一个确定的依赖性信息，将服务集群中的服务中的第一个服务确定为性能数据模式的根本原因。

公开(公告)号：US09582395B2

公开(公告)日：2017-02-28

申请号：US13826942

申请日：2013-03-14

Applicant: Netflix, Inc.

Inventor： Philip Simon Tuffs ,  Roy Rapoport ,  Ariel Tseitlin

IPC: G06F11/00 ,  G06F11/34 ,  G06F11/07

CPC classification number: G06F11/3452 ,  G06F11/0709 ,  G06F11/079 ,  G06F11/3409 ,  G06F11/3495

Abstract: Techniques are described for identifying a root cause of a pattern of performance data in a system including a plurality of services. Embodiments provide dependency information for each of the plurality of services, where at least one of the plurality of services is dependent upon a first one of the plurality of services. Each of the plurality of services is monitored to collect performance data for the respective service. Embodiments further analyze the performance data to identify a cluster of services that each follow a pattern of performance data. The first one of the services in the cluster of services is determined to be a root cause of the pattern of performance data, based on the determined dependency information for each of the plurality of services.

Abstract translation: 描述了用于识别包括多个服务的系统中的性能数据模式的根本原因的技术。 实施例为多个服务中的每一个提供依赖信息，其中多个服务中的至少一个服务依赖于多个服务中的第一个服务。 监视多个服务中的每一个以收集相应服务的性能数据。 实施例进一步分析性能数据，以识别每个遵循性能数据模式的服务集群。 基于针对多个服务中的每一个确定的依赖性信息，将服务集群中的服务中的第一个服务确定为性能数据模式的根本原因。

公开(公告)号：US20150113120A1

公开(公告)日：2015-04-23

申请号：US14057898

申请日：2013-10-18

Applicant: Netflix, Inc.

Inventor： Daniel Isaac Jacobson ,  Neeraj Joshi ,  Puneet Oberai ,  Yong Yuan ,  Philip Simon Tuffs

IPC: G06N5/04 ,  H04L12/911

Abstract: Techniques for predictively scaling a distributed application are described. Embodiments could monitor performance of an application within a cloud computing environment over a first window of time to collect historical performance data. Here, the application comprises a plurality of application instances. A workload of the application could be monitored over a second window of time to collect historical workload data. Embodiments could analyze both the historical performance data and the historical workload data to determine one or more scaling patterns for the application. Upon determining a present state of the application matches one of the one or more scaling patterns, a plan for predictively scaling the application could be determined. Embodiments could then predictively scale the plurality of application instances, based on the determined plan.

Abstract translation: 描述了用于预测性地缩放分布式应用的技术。 实施例可以在第一时间窗口内监视云计算环境内的应用程序的性能，以收集历史性能数据。 这里，应用程序包括多个应用程序实例。 可以在第二个时间窗口监视应用程序的工作负载，以收集历史工作负载数据。 实施例可以分析历史性能数据和历史工作负载数据，以确定应用的一个或多个缩放模式。 在确定应用程序的当前状态与一个或多个缩放模式中的一个匹配时，可以确定用于预测缩放应用程序的计划。 然后，实施例可以基于所确定的计划来预测性地缩放多个应用程序实例。

公开(公告)号：US20140282422A1

公开(公告)日：2014-09-18

申请号：US13796923

申请日：2013-03-12

Applicant: NETFLIX, INC.

Inventor： Philip Simon Tuffs ,  Roy Rapoport ,  Ariel Tseitlin

IPC: G06F11/34

CPC classification number: G06F11/3452 ,  G06F11/3428 ,  G06F2201/865

Abstract: Techniques for evaluating a second version of software. Embodiments selectively route incoming requests to software instances within a plurality of baseline instances and a plurality of canary instances, where the baseline instances run a first software version and the canary instances run the second software version. The software instances are monitored to collect performance data for a plurality of performance metrics. Embodiments calculate aggregate baseline performance metrics, where each of the aggregate baseline performance metrics is calculated based on the collected performance data for the plurality of baseline instances. For each of the performance metrics and canary instances, embodiments calculate a relative performance value that measures the collected performance data for the respective canary instance and for the respective performance metric, relative to the corresponding aggregate baseline performance metric. A final measure of performance is calculated for the second version of software, based on the relative performance values.

Abstract translation: 用于评估第二版软件的技术。 实体选择性地将传入请求路由到多个基线实例和多个金丝雀实例中的软件实例，其中基准实例运行第一软件版本，而金丝雀实例运行第二软件版本。 监视软件实例以收集多个性能度量的性能数据。 实施例计算聚合基线性能度量，其中基于多个基准实例的收集的性能数据来计算每个聚合基线性能指标。 对于每个性能指标和金丝雀实例，实施例计算相对性能值，该相对性能值相对于相应的聚合基线性能度量来衡量针对相应的金丝雀实例和相应的性能度量的收集的性能数据。 基于相对性能值，针对第二版软件计算性能的最终测量。

公开(公告)号：US10552745B2

公开(公告)日：2020-02-04

申请号：US14057898

申请日：2013-10-18

Applicant: Netflix, Inc.

Inventor： Daniel Isaac Jacobson ,  Neeraj Joshi ,  Puneet Oberai ,  Yong Yuan ,  Philip Simon Tuffs

IPC: H04L12/26 ,  G06N5/04 ,  H04L12/911

Abstract: Techniques for predictively scaling a distributed application are described. Embodiments could monitor performance of an application within a cloud computing environment over a first window of time to collect historical performance data. Here, the application comprises a plurality of application instances. A workload of the application could be monitored over a second window of time to collect historical workload data. Embodiments could analyze both the historical performance data and the historical workload data to determine one or more scaling patterns for the application. Upon determining a present state of the application matches one of the one or more scaling patterns, a plan for predictively scaling the application could be determined. Embodiments could then predictively scale the plurality of application instances, based on the determined plan.

公开(公告)号：US10318399B2

公开(公告)日：2019-06-11

申请号：US13796923

申请日：2013-03-12

Applicant: Netflix, Inc.

Inventor： Philip Simon Tuffs ,  Roy Rapoport ,  Ariel Tseitlin

IPC: G06F9/44 ,  G06F11/34

Abstract: Techniques for evaluating a second version of software. Embodiments selectively route incoming requests to software instances within a plurality of baseline instances and a plurality of canary instances, where the baseline instances run a first software version and the canary instances run the second software version. The software instances are monitored to collect performance data for a plurality of performance metrics. Embodiments calculate aggregate baseline performance metrics, where each of the aggregate baseline performance metrics is calculated based on the collected performance data for the plurality of baseline instances. For each of the performance metrics and canary instances, embodiments calculate a relative performance value that measures the collected performance data for the respective canary instance and for the respective performance metric, relative to the corresponding aggregate baseline performance metric. A final measure of performance is calculated for the second version of software, based on the relative performance values.