公开(公告)号：US09406029B2

公开(公告)日：2016-08-02

申请号：US14143012

申请日：2013-12-30

Applicant: NetApp, Inc.

Inventor： Sai Rama Krishna Susarla ,  Kaladhar Voruganti ,  Vipul Mathur

IPC: G06F17/30 ,  G06N99/00 ,  G06F11/34

CPC classification number: G06N99/005 ,  G06F11/3409 ,  G06F17/30294 ,  G06F17/30587

Abstract: Described herein is a system and method for dynamically managing service-level objectives (SLOs) for workloads of a cluster storage system. Proposed states/solutions of the cluster may be produced and evaluated to select one that achieves the SLOs for each workload. A planner engine may produce a state tree comprising nodes, each node representing a proposed state/solution. New nodes may be added to the state tree based on new solution types that are permitted, or nodes may be removed based on a received time constraint for executing a proposed solution or a client certification of a solution. The planner engine may call an evaluation engine to evaluate proposed states, the evaluation engine using an evaluation function that considers SLO, cost, and optimization goal characteristics to produce a single evaluation value for each proposed state. The planner engine may call a modeler engine that is trained using machine learning techniques.

公开(公告)号：US20160170794A1

公开(公告)日：2016-06-16

申请号：US15048907

申请日：2016-02-19

Applicant: NetApp, Inc.

Inventor： Lakshmi Narayanan Bairavasundaram ,  Gokul Soundararajan ,  Vipul Mathur ,  Kaladhar Voruganti

IPC: G06F9/455 ,  G06F9/50 ,  H04L29/08

CPC classification number: G06F9/45558 ,  G06F9/5016 ,  G06F9/5077 ,  G06F9/5083 ,  G06F11/073 ,  G06F11/3409 ,  G06F12/0806 ,  G06F2009/4557 ,  G06F2009/45583 ,  G06F2009/45595 ,  H04L41/0816 ,  H04L41/0896 ,  H04L67/2842

Abstract: It is detected that a metric associated with a first workload has breached a first threshold. It is determined that the first workload and a second workload access the same storage resources, wherein the storage resources are associated with a storage server. It is determined that the metric is impacted by the first workload and the second workload accessing the same storage resources. A candidate solution is identifier. An estimated impact of a residual workload is determined based, at least in part, on the candidate solution. A level of caching of at least one of the first workload or the second workload is adjusted based, at least in part, on the estimated impact of the residual workload.

Abstract translation: 检测到与第一工作负荷相关联的度量已经违反了第一阈值。 确定第一工作负载和第二工作负载访问相同的存储资源，其中存储资源与存储服务器相关联。 确定该度量受到第一工作负载的影响，并且第二工作负载访问相同的存储资源。 候选解决方案是标识符。 至少部分地基于候选解决方案确定剩余工作量的估计影响。 至少部分地基于剩余工作量的估计影响来调整至少一个第一工作负载或第二工作负载的高速缓存。

公开(公告)号：US09336489B2

公开(公告)日：2016-05-10

申请号：US14513044

申请日：2014-10-13

Applicant: NetApp, Inc.

Inventor： Rukma Talwadker ,  Kaladhar Voruganti ,  David Slik

IPC: G05B13/04 ,  G06N7/04 ,  H04L29/08 ,  G06N5/04

CPC classification number: G06N5/048 ,  G05B13/04 ,  H04L67/10

Abstract: In the area of storage management, service automation can be realized through the use of “MAPE” loop(s). A Planner (P) interacts with the Monitoring (M), Analysis (A) and Execution (E) components in a closed loop. For each new option or potential planning action the Planner (P) invokes the Analysis (A) component. The correctness, as well as effectiveness, of the planning decision is dependent on the Analysis (A) component. Embodiments can utilize an adaptive Analysis (A) component (i.e., an analysis component that can be retrained) that also associates a value of confidence and a corresponding error in the evaluation along with a predicted impact. The Planner (P) component uses this additional information for quoting the final impact of a particular planning action as part of an adaptive MAPE loop to provide improved resource utilization and resource management.

Abstract translation: 在存储管理方面，可以通过使用“MAPE”循环实现业务自动化。 计划员（P）在闭环中与监控（M），分析（A）和执行（E）组件进行交互。 对于每个新选项或潜在的计划操作，计划员（P）调用Analysis（A）组件。 规划决策的正确性和有效性取决于分析（A）组成部分。 实施例可以利用自适应分析（A）分量（即，可再培训的分析组件），其也可以将评估中的置信度和相应误差与预测的影响相关联。 计划员（P）组件使用这些附加信息来引用特定计划动作的最终影响作为自适应MAPE循环的一部分，以提供改进的资源利用和资源管理。

公开(公告)号：US09208181B2

公开(公告)日：2015-12-08

申请号：US14523801

申请日：2014-10-24

Applicant: NetApp Inc.

Inventor： Vineet Pandey ,  Chhavi Sharma ,  Ranjit Kumar ,  Parag Deshmukh ,  Kaladhar Voruganti

IPC: G06F17/30 ,  G06F9/455 ,  G06F15/173 ,  H04L29/08

CPC classification number: G06F17/303 ,  G06F9/45533 ,  G06F15/17331 ,  G06F17/30079 ,  G06F17/30088 ,  G06F17/30179 ,  G06F17/30607 ,  H04L67/06 ,  H04L67/1023 ,  H04L67/1097

Abstract: One or more techniques and/or systems are provided for migrating a dataset from a file storage system to an object storage system. That is, a snapshot of a file system may be received from the file storage system. The snapshot may comprise file data associated with a file of the file system. The file may be converted into an object using the file data. The object may be stored within a data constituent volume of the object storage system. A namespace volume, used to track objects, may be populated with a redirector that maps a front-end data path (e.g., a path used by clients to reference the object) to a back-end data path that specifies a path to the object within the data constituent volume. In this way, a dataset of one or more files may be migrated from the file storage system to the object storage system.

Abstract translation: 提供一个或多个技术和/或系统用于将数据集从文件存储系统迁移到对象存储系统。 也就是说，可以从文件存储系统接收文件系统的快照。 快照可以包括与文件系统的文件相关联的文件数据。 可以使用文件数据将文件转换成对象。 对象可以存储在对象存储系统的数据构成卷内。 用于跟踪对象的命名空间卷可以用映射前端数据路径（例如，客户端引用对象的路径）的重定向器填充到指定对象的路径的后端数据路径 在数据构成卷内。 以这种方式，一个或多个文件的数据集可以从文件存储系统迁移到对象存储系统。

公开(公告)号：US20150142942A1

公开(公告)日：2015-05-21

申请号：US14338157

申请日：2014-07-22

Applicant: NetApp, Inc.

Inventor： Kaladhar Voruganti ,  Steven R. Kleiman ,  James Hartwell Holl, II ,  Gokul Soundararajan ,  Shailaja Kamila ,  Subra Mohan

IPC: H04L12/24 ,  G06F17/30 ,  G06F11/14 ,  H04L29/08

CPC classification number: G06F11/1446 ,  G06F11/1448 ,  G06F11/1458 ,  G06F11/1461 ,  G06F11/2094 ,  H04L41/5096 ,  H04L67/1097 ,  H04L67/28

Abstract: Technology is disclosed for managing network storage services by service level objectives (SLOs). The method receives multiple service level capability (SLC) templates; creates at least one storage service level (SSL) instance using at least one of the SLC templates; provisions a storage object located in a network storage infrastructure based on the SSL instance; and services storage requests using the storage object.

Abstract translation: 披露了按服务级目标（SLO）管理网络存储服务的技术。 该方法接收多个服务级别的能力（SLC）模板; 使用至少一个SLC模板创建至少一个存储服务级别（SSL）实例; 基于SSL实例提供位于网络存储基础设施中的存储对象; 以及使用存储对象的服务存储请求。

公开(公告)号：US20150046654A1

公开(公告)日：2015-02-12

申请号：US14523655

申请日：2014-10-24

Applicant: NetApp, Inc.

Inventor： Gokul Soundararajan ,  Kaladhar Voruganti ,  Lakshmi Narayanan Bairavasunda ,  Priya Sehgal ,  Vipul Mathur

IPC: G06F12/08

CPC classification number: G06F12/0811 ,  G06F12/0813 ,  G06F12/0888 ,  G06F2212/2515 ,  G06F2212/284 ,  G06F2212/301 ,  G06F2212/601

Abstract: A change in workload characteristics detected at one tier of a multi-tiered cache is communicated to another tier of the multi-tiered cache. Multiple caching elements exist at different tiers, and at least one tier includes a cache element that is dynamically resizable. The communicated change in workload characteristics causes the receiving tier to adjust at least one aspect of cache performance in the multi-tiered cache. In one aspect, at least one dynamically resizable element in the multi-tiered cache is resized responsive to the change in workload characteristics.

Abstract translation: 在多层缓存的一层检测到的工作负载特性的改变被传送到多层高速缓存的另一层。 多个缓存元素存在于不同的层，并且至少一个层包括可动态调整大小的高速缓存元素。 所传达的工作负载特性的改变使得接收层在多层缓存中调整缓存性能的至少一个方面。 在一个方面，响应于工作负载特性的变化来调整多层缓存中的至少一个可动态调整大小的元素。

公开(公告)号：US20150032683A1

公开(公告)日：2015-01-29

申请号：US14513044

申请日：2014-10-13

Applicant: NetApp, Inc.

Inventor： Rukma Talwadker ,  Kaladhar Voruganti ,  David Slik

IPC: G06N5/04 ,  H04L29/08

CPC classification number: G06N5/048 ,  G05B13/04 ,  H04L67/10

Abstract: In the area of storage management, service automation can be realized through the use of “MAPE” loop(s). A Planner (P) interacts with the Monitoring (M), Analysis (A) and Execution (E) components in a closed loop. For each new option or potential planning action the Planner (P) invokes the Analysis (A) component. The correctness, as well as effectiveness, of the planning decision is dependent on the Analysis (A) component. Embodiments can utilize an adaptive Analysis (A) component (i.e., an analysis component that can be retrained) that also associates a value of confidence and a corresponding error in the evaluation along with a predicted impact. The Planner (P) component uses this additional information for quoting the final impact of a particular planning action as part of an adaptive MAPE loop to provide improved resource utilization and resource management.

Abstract translation: 在存储管理方面，可以通过使用“MAPE”循环实现业务自动化。 计划员（P）在闭环中与监控（M），分析（A）和执行（E）组件进行交互。 对于每个新选项或潜在的计划操作，计划员（P）调用Analysis（A）组件。 规划决策的正确性和有效性取决于分析（A）组成部分。 实施例可以利用自适应分析（A）分量（即，可再培训的分析组件），其也可以将评估中的置信度和相应误差与预测的影响相关联。 计划员（P）组件使用这些附加信息来引用特定计划动作的最终影响作为自适应MAPE循环的一部分，以提供改进的资源利用和资源管理。

公开(公告)号：US20140380332A1

公开(公告)日：2014-12-25

申请号：US14484780

申请日：2014-09-12

Applicant: NETAPP, INC.

Inventor： Vipul Mathur ,  Neeraja Yadwadkar ,  Lakshmi Narayanan Bairavasundaram ,  Rukma Ameet Talwadker ,  Kaladhar Voruganti ,  Sai Rama Krishna Susaria

IPC: G06F9/46

CPC classification number: H04L67/1008 ,  G06F9/465 ,  G06F11/3006 ,  G06F11/3034 ,  G06F11/3409 ,  G06F11/3442 ,  G06F11/3447 ,  G06F11/3485 ,  G06F2201/81 ,  G06F2201/86 ,  H04L12/54 ,  H04L29/06 ,  H04L41/5009 ,  H04L47/70 ,  H04L67/1097 ,  Y02D10/34

Abstract: Described herein is a system and method for dynamically managing service-level objectives (SLOs) for workloads of a cluster storage system. Proposed states/solutions of the cluster may be produced and evaluated to select one that achieves the SLOs for each workload. A planner engine may produce a state tree comprising nodes, each node representing a proposed state/solution. New nodes may be added to the state tree based on new solution types that are permitted, or nodes may be removed based on a received time constraint for executing a proposed solution or a client certification of a solution. The planner engine may call an evaluation engine to evaluate proposed states, the evaluation engine using an evaluation function that considers SLO, cost, and optimization goal characteristics to produce a single evaluation value for each proposed state. The planner engine may call a modeler engine that is trained using machine learning techniques.

Abstract translation: 这里描述了用于动态管理用于集群存储系统的工作负载的服务级目标（SLO）的系统和方法。 可以生成和评估集群的建议状态/解决方案，以选择为每个工作负载实现SLO的状态/解决方案。 计划器引擎可以产生包括节点的状态树，每个节点表示提出的状态/解。 可以基于允许的新解决方案类型将新节点添加到状态树，或者可以基于接收到的时间约束来移除节点，以执行解决方案或解决方案的客户端认证。 计划器引擎可以调用评估引擎来评估提出的状态，评估引擎使用考虑SLO，成本和优化目标特征的评估函数，以产生每个建议状态的单个评估值。 计划器引擎可以调用使用机器学习技术训练的建模者引擎。

公开(公告)号：US20140244643A1

公开(公告)日：2014-08-28

申请号：US13781619

申请日：2013-02-28

Applicant: NetApp, Inc.

Inventor： Jayanta Basak ,  Kushal Wadhwani ,  Kaladhar Voruganti

IPC: G06F17/30

CPC classification number: G06F16/355 ,  G06F11/3409 ,  G06F11/3414 ,  G06F11/3419 ,  G06F16/217 ,  G06F2201/83

Abstract: An embodiment of the invention provides an apparatus and method for classifying a workload of a computing entity. In an embodiment, the computing entity samples a plurality of values for a plurality of parameters of the workload. Based on the plurality of values of each parameter, the computing entity determines a parameter from the plurality of parameters that the computing entity's response time is dependent on. Here, the computing entity's response time is indicative of a time required by the computing entity to respond to a service request from the workload. Further, based on the identified significant parameter, the computing entity classifies the workload of the computing entity by selecting a workload classification from a plurality of predefined workload classifications.

Abstract translation: 本发明的实施例提供了一种用于对计算实体的工作负载进行分类的装置和方法。 在一个实施例中，计算实体针对工作负载的多个参数对多个值进行采样。 基于每个参数的多个值，计算实体根据计算实体的响应时间依赖于多个参数确定参数。 这里，计算实体的响应时间表示计算实体响应来自工作负载的服务请求所需的时间。 此外，基于所识别的重要参数，计算实体通过从多个预定义的工作负载分类中选择工作负载分类来对计算实体的工作负载进行分类。

公开(公告)号：US20140223014A1

公开(公告)日：2014-08-07

申请号：US14087372

申请日：2013-11-22

Applicant: NetApp, Inc.

Inventor： Lakshmi Narayanan Bairavasundaram ,  Gokul Soundararajan ,  Vipul Mathur ,  Kaladhar Voruganti ,  Darren Sawyer

IPC: H04L29/08

CPC classification number: H04L67/1023 ,  G06F3/0605 ,  G06F3/0683 ,  G06F9/5011 ,  G06F11/2094 ,  G06F11/3409 ,  G06F11/3442 ,  G06F11/3495 ,  G06F2201/81 ,  H04L41/5009 ,  H04L49/505 ,  H04L67/1097 ,  H04L69/40 ,  H04L2012/5631 ,  H04L2012/5636

Abstract: Collaborative management of shared resources is implemented by a storage server receiving, from a first resource manager, notification of a violation for a service provided by the storage server or device coupled to the storage server. The storage server further receives, from each of a plurality of resource managers, an estimated cost of taking a corrective action to mitigate the violation and selects a corrective action proposed by one of the plurality of resource managers based upon the estimated cost. The storage server directs the resource manager that proposed the selected corrective action to perform the selected corrective action.

Abstract translation: 共享资源的协同管理由存储服务器实现，存储服务器从第一资源管理器接收由与存储服务器耦合的存储服务器或设备提供的服务的违规的通知。 存储服务器还从多个资源管理器中的每一个进一步接收采取纠正措施以减轻违规的估计成本，并且基于估计的成本选择由多个资源管理者之一提出的纠正措施。 存储服务器指示提出所选修正操作的资源管理器执行所选的纠正措施。