专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US07653668B1 Fault tolerant multi-stage data replication with relaxed coherency guarantees 有权
标题翻译：具有轻松一致性保证的容错多级数据复制
公开(公告)号：US07653668B1
公开(公告)日：2010-01-26
申请号：US11286619
申请日：2005-11-23
申请人： Radha Shelat , Dilip Madhusudan Ranade , Navin Kabra , BijayaLaxmi Nanda , Anindya Banerjee
发明人： Radha Shelat , Dilip Madhusudan Ranade , Navin Kabra , BijayaLaxmi Nanda , Anindya Banerjee
IPC分类号： G06F17/30
CPC分类号： G06F17/30212 , G06F11/2094
摘要： A system for fault tolerant multi-stage data replication with relaxed coherency comprises one or more processors and memory coupled to the processors, where the memory stores program instructions executable by the processors to implement a data replication service. In response to an update request directed at a data object, in a first stage of replication, the data replication service is configured to update respective primary replicas at a plurality of primary replica sites for the data object, and to store metadata descriptors at each of the plurality of primary replica sites. The data replication service is configured to determine whether a notification of the update is to be sent to any other replica sites. In response to identifying such other replica sites, an update notification including contents of a metadata descriptor may be sent to the other replica sites in one or more additional replication stages.
摘要翻译：具有轻松一致性的容错多级数据复制的系统包括耦合到处理器的一个或多个处理器和存储器，其中存储器存储可由处理器执行的程序指令以实现数据复制服务。响应于针对数据对象的更新请求，在第一复制阶段中，数据复制服务被配置为更新数据对象的多个主复本站点处的各个主要副本，并且在每个复制站点存储元数据描述符多个主要副本站点。配置数据复制服务以确定是否将更新的通知发送到任何其他副本站点。响应于识别这样的其他复制站点，可以在一个或多个附加复制阶段将包括元数据描述符的内容的更新通知发送到其他副本站点。

2. 发明授权

US07831735B1 Coherency of replicas for a distributed file sharing system 有权
标题翻译：一个分布式文件共享系统的副本的一致性
公开(公告)号：US07831735B1
公开(公告)日：2010-11-09
申请号：US12396280
申请日：2009-03-02
申请人： Navin Kabra , Anindya Banerjee , BijayaLaxmi Nanda , Sivaramakrishna Ramadugu Venkata , Dilip Madhusudan Ranade , Radha Shelat , Laxmikant Vithal Gunda
发明人： Navin Kabra , Anindya Banerjee , BijayaLaxmi Nanda , Sivaramakrishna Ramadugu Venkata , Dilip Madhusudan Ranade , Radha Shelat , Laxmikant Vithal Gunda
IPC分类号： G06F15/16 , G06F7/00
CPC分类号： G06F17/30212 , G06F17/30165 , G06F17/30176 , G06F17/30227 , Y10S707/99952
摘要： A plurality of data objects may be replicated across a plurality of computing nodes coupled to a network. The network may include a first node operable to initiate an update operation to update a plurality of replicas of a first object. If one or more of the replicas are not reachable then the update operation may update a subset (e.g., a quorum) but not all of the replicas. For each node on which one of the replicas was updated in the update operation, the node may add the object to a list of incoherent objects. The list of incoherent objects may subsequently be used to bring the lagging replicas in sync with the replicas that were updated. In another embodiment, a plurality of replicas of an object may be stored on a plurality of nodes, similarly as described above. A first node that stores a replica of the object may store a first timestamp associated with the replica on the first node. The timestamp may be used to ensure that the replica on the first node is coherent with respect to one or more other replicas by periodically communicating with the one or more other replicas when a threshold amount of time has passed without the replica on the first node receiving an update.
摘要翻译：可以跨耦合到网络的多个计算节点复制多个数据对象。网络可以包括可操作以发起更新操作以更新第一对象的多个副本的第一节点。如果一个或多个副本不可达，则更新操作可以更新子集（例如，法定人数），但不能更新所有副本。对于在更新操作中更新其中一个副本的每个节点，节点可以将对象添加到不相干对象的列表中。随后可以使用非相干对象的列表将滞后副本与已更新的副本同步。在另一个实施例中，类似于上述，可以将多个对象的副本存储在多个节点上。存储对象的副本的第一节点可以在第一节点上存储与副本相关联的第一时间戳。时间戳可以用于通过在经过第一节点接收到副本的阈值时间段之后通过周期性地与一个或多个其他副本进行通信来确保第一节点上的副本相对于一个或多个其他副本是一致的更新。

3. 发明授权

US07500020B1 Coherency of replicas for a distributed file sharing system 有权
标题翻译：一个分布式文件共享系统的副本的一致性
公开(公告)号：US07500020B1
公开(公告)日：2009-03-03
申请号：US10749365
申请日：2003-12-31
申请人： Navin Kabra , Anindya Banerjee , BijayaLaxmi Nanda , Sivaramakrishna Ramadugu Venkata , Dilip Madhusudan Ranade , Radha Shelat , Laxmikant Vithal Gunda
发明人： Navin Kabra , Anindya Banerjee , BijayaLaxmi Nanda , Sivaramakrishna Ramadugu Venkata , Dilip Madhusudan Ranade , Radha Shelat , Laxmikant Vithal Gunda
IPC分类号： G06F15/16 , G06F12/00
CPC分类号： G06F17/30212 , G06F17/30165 , G06F17/30176 , G06F17/30227 , Y10S707/99952
摘要： A plurality of data objects may be replicated across a plurality of computing nodes coupled to a network. The network may include a first node operable to initiate an update operation to update a plurality of replicas of a first object. If one or more of the replicas are not reachable then the update operation may update a subset (e.g., a quorum) but not all of the replicas. For each node on which one of the replicas was updated in the update operation, the node may add the object to a list of incoherent objects. The list of incoherent objects may subsequently be used to bring the lagging replicas in sync with the replicas that were updated. In another embodiment, a plurality of replicas of an object may be stored on a plurality of nodes, similarly as described above. A first node that stores a replica of the object may store a first timestamp associated with the replica on the first node. The timestamp may be used to ensure that the replica on the first node is coherent with respect to one or more other replicas by periodically communicating with the one or more other replicas when a threshold amount of time has passed without the replica on the first node receiving an update.
摘要翻译：可以跨耦合到网络的多个计算节点复制多个数据对象。网络可以包括可操作以发起更新操作以更新第一对象的多个副本的第一节点。如果一个或多个副本不可达，则更新操作可以更新子集（例如，法定人数），但不能更新所有副本。对于在更新操作中更新其中一个副本的每个节点，节点可以将对象添加到不相干对象的列表中。随后可以使用非相干对象的列表将滞后副本与已更新的副本同步。在另一个实施例中，类似于上述，可以将多个对象的副本存储在多个节点上。存储对象的副本的第一节点可以在第一节点上存储与副本相关联的第一时间戳。时间戳可以用于通过在经过第一节点接收到副本的阈值时间段之后通过周期性地与一个或多个其他副本进行通信来确保第一节点上的副本相对于一个或多个其他副本是一致的更新。

4. 发明授权

US07734820B1 Adaptive caching for a distributed file sharing system 有权
标题翻译：分布式文件共享系统的自适应缓存
公开(公告)号：US07734820B1
公开(公告)日：2010-06-08
申请号：US10750426
申请日：2003-12-31
申请人： Dilip Madhusudan Ranade , Radha Shelat , Navin Kabra
发明人： Dilip Madhusudan Ranade , Radha Shelat , Navin Kabra
IPC分类号： G06F15/16
CPC分类号： G06F17/30067
摘要： A system and method for adaptively responding to update messages for a replica of a data object. A node may maintain first information regarding a replica of a data object. In various embodiments, the first information may comprise any kind of information regarding the replica of the data object. In one embodiment, the first information may comprise information indicative of accesses to the replica of the data object, e.g., may comprise access history information. In response to receiving a message for updating the replica of the data object, the node may respond to the message based on the first information. For example, the node may select one or more operations to perform based on the first information and may perform the one or more selected operations.
摘要翻译：一种用于自适应地响应数据对象的副本的更新消息的系统和方法。节点可以维护关于数据对象的副本的第一信息。在各种实施例中，第一信息可以包括关于数据对象的副本的任何种类的信息。在一个实施例中，第一信息可以包括指示对数据对象的副本的访问的信息，例如可以包括访问历史信息。响应于接收到用于更新数据对象的副本的消息，节点可以基于第一信息来响应消息。例如，节点可以基于第一信息选择要执行的一个或多个操作，并且可以执行一个或多个所选择的操作。

5. 发明授权

US07698414B1 Global monitoring and statistics collection for a plurality of computing nodes 有权
标题翻译：用于多个计算节点的全局监视和统计信息收集
公开(公告)号：US07698414B1
公开(公告)日：2010-04-13
申请号：US10855603
申请日：2004-05-27
申请人： Radha Shelat , Sivaramakrishna Ramadugu Venkata , Navin Kabra
发明人： Radha Shelat , Sivaramakrishna Ramadugu Venkata , Navin Kabra
IPC分类号： G06F15/173 , G06F17/30
CPC分类号： G06F17/30209 , H04L29/12783 , H04L61/35
摘要： A system including a plurality of nodes operable to exchange information using location-independent addresses. According to one embodiment of a method for obtaining information in a network of nodes, one or more nodes may each create an instance of a location-independent address A. The location-independent address A may represent an address for sending request messages (also referred to simply as requests) in order to obtain information or data from the one or more nodes with which the location-independent address A is associated (i.e., from the one or more nodes that create instances of the location-independent address A). According to one embodiment of a method for storing information in a network of nodes, one or more nodes may each create an instance of a location-independent address A, similarly as described above. However, instead of representing an address for sending requests to obtain information or data, in this case the location-independent address A may represent an address for sending requests to store information or data.
摘要翻译：一种包括可操作以使用与位置无关的地址来交换信息的多个节点的系统。根据用于在节点网络中获得信息的方法的一个实施例，一个或多个节点可以各自创建与位置无关的地址A的实例。位置无关地址A可以表示用于发送请求消息的地址（也称为以简单地作为请求），以便从与位置无关地址A相关联的一个或多个节点（即，来自创建与位置无关的地址A的实例的一个或多个节点）获得信息或数据。根据用于在节点网络中存储信息的方法的一个实施例，类似于上述，一个或多个节点可以各自创建与位置无关的地址A的实例。然而，代替用于发送请求以获取信息或数据的地址，在这种情况下，位置无关地址A可以表示用于发送存储信息或数据的请求的地址。

6. 发明授权

US08332690B1 Method and apparatus for managing failures in a datacenter 有权
标题翻译：管理数据中心故障的方法和装置
公开(公告)号：US08332690B1
公开(公告)日：2012-12-11
申请号：US12147767
申请日：2008-06-27
申请人： Anindya Banerjee , Bijaya Laxmi Nanda , Navin Kabra
发明人： Anindya Banerjee , Bijaya Laxmi Nanda , Navin Kabra
IPC分类号： G06F11/00
CPC分类号： G06F11/0727 , G06F11/0793
摘要： A method and apparatus for providing information to mitigate a current failure in a datacenter is described. The method comprises processing historical data regarding at least one failure. The historical data includes at least one action related to the at least one failure. The method further includes determining information for mitigating the current failure based on the historical data.
摘要翻译：描述了一种用于提供信息以减轻数据中心中的当前故障的方法和装置。该方法包括处理关于至少一个故障的历史数据。历史数据包括与至少一个故障有关的至少一个动作。该方法还包括基于历史数据确定用于减轻当前故障的信息。

7. 发明授权

US09069992B1 System and method for reducing data loss prevention scans 有权
标题翻译：减少数据丢失防护扫描的系统和方法
公开(公告)号：US09069992B1
公开(公告)日：2015-06-30
申请号：US12533932
申请日：2009-07-31
申请人： Amol Vaikar , Anindya Banerjee , Navin Kabra , Amit Verma
发明人： Amol Vaikar , Anindya Banerjee , Navin Kabra , Amit Verma
IPC分类号： G06F21/10 , G06F21/88
CPC分类号： G06F21/88 , G06F21/6218
摘要： A computing device receives a file. The computing device determines whether the file has previously been scanned for violations of a data loss prevention policy. If the file was previously scanned, a result of the previous scan is used to decide whether or not the file violates the data loss prevention policy. If the file was not previously scanned, the file is scanned to decide whether or not the file violates the data loss prevention policy.
摘要翻译：计算设备接收文件。计算设备确定是否先前扫描了文件是否违反数据丢失防护策略。如果先前扫描了文件，则使用上一次扫描的结果来决定该文件是否违反了数据丢失保护策略。如果以前未扫描该文件，则会扫描文件以确定该文件是否违反了数据丢失保护策略。

8. 发明授权

US07478400B1 Efficient distributed transaction protocol for a distributed file sharing system 有权
标题翻译：分布式文件共享系统的高效分布式事务协议
公开(公告)号：US07478400B1
公开(公告)日：2009-01-13
申请号：US10749300
申请日：2003-12-31
申请人： Anindya Banerjee , Sivaramakrishna Ramadugu Venkata , Navin Kabra , Kai Wong , Bijaya Laxmi Nanda
发明人： Anindya Banerjee , Sivaramakrishna Ramadugu Venkata , Navin Kabra , Kai Wong , Bijaya Laxmi Nanda
IPC分类号： G06F3/00 , G06F15/16
CPC分类号： G06F17/30067
摘要： An efficient distributed transaction protocol for a distributed file sharing system is described. One embodiment of the protocol involves a transaction coordinator node sending three messages to participant nodes, where the participant nodes reply to only two of those messages. In one embodiment, the transaction coordinator node may return success to the client application after receiving replies to the second message from the participant nodes. The protocol may provide benefits associated with traditional three-phase commit (3PC) protocols while also decreasing response time latencies experienced by client applications and decreasing network traffic.
摘要翻译：描述了一种用于分布式文件共享系统的高效分布式事务协议。该协议的一个实施例涉及事务协调器节点向参与者节点发送三个消息，其中参与者节点只回复这些消息中的两个。在一个实施例中，事务协调器节点可以在从参与者节点接收到第二消息的答复之后将成功返回给客户端应用。该协议可以提供与传统三阶段提交（3PC）协议相关联的优点，同时还减少客户端应用所经历的响应时间延迟并减少网络流量。

9. 发明授权

US09508044B1 Method and apparatus for managing configurations 有权
标题翻译：用于管理配置的方法和装置
公开(公告)号：US09508044B1
公开(公告)日：2016-11-29
申请号：US13344761
申请日：2012-01-06
申请人： Venkeepuram Satish , Navin Kabra , Subhojit Roy , Himanshu Ashwani
发明人： Venkeepuram Satish , Navin Kabra , Subhojit Roy , Himanshu Ashwani
IPC分类号： G06E1/00 , G06E3/00 , G06F15/18 , G06G7/00 , G06N99/00 , G06N3/08 , G06N3/04
CPC分类号： G06N99/005 , G06N3/04 , G06N3/08 , G06N7/005 , G06Q10/0631
摘要： A method and apparatus for managing configurations of computer resources in a datacenter is described. In one embodiment, a method comprises analyzing multiple configurations using rule information to produce an analysis result where each configuration in the multiple configurations defines a configuration of a resource that is managed by the data center, training a Bayesian classifier using the analysis result, and classifiying a second configuration using the trained Bayesian classifier.
摘要翻译：描述了一种用于管理数据中心中的计算机资源的配置的方法和装置。在一个实施例中，一种方法包括使用规则信息来分析多个配置以产生分析结果，其中多个配置中的每个配置定义由数据中心管理的资源的配置，使用分析结果训练贝叶斯分类器，并分类使用训练有素的贝叶斯分类器的第二种配置。

10. 发明授权

US06732084B1 Method and apparatus for parallel execution of trigger actions 有权
标题翻译：用于并行执行触发动作的方法和装置
公开(公告)号：US06732084B1
公开(公告)日：2004-05-04
申请号：US09470227
申请日：1999-12-22
申请人： Navin Kabra , Jignesh M. Patel , Jie-Bing Yu , Biswadeep Nag , Jian-Jun Chen
发明人： Navin Kabra , Jignesh M. Patel , Jie-Bing Yu , Biswadeep Nag , Jian-Jun Chen
IPC分类号： G06F1730
CPC分类号： G06F17/3051 , Y10S707/99932 , Y10S707/99933
摘要： A method and apparatus for parallel execution of trigger actions. One or more embodiments of the invention comprise providing a C++ class (hereinafter referred to as “dispatcher”) that can take an SQL query or trigger action and start parallel execution of the trigger action. The trigger action is optimized and parallelized. The dispatcher executes the trigger action, sets up the communication links between the various operators in the trigger action, and ensures that all the results are sent back to the trigger.
摘要翻译：一种用于并行执行触发动作的方法和装置。本发明的一个或多个实施例包括提供可以采取SQL查询或触发动作并开始并行执行触发动作的C ++类（以下称为“调度器”）。触发动作被优化和并行化。调度程序执行触发动作，在触发动作中设置各种运算符之间的通信链路，并确保所有结果都发送回触发器。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式