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. 发明申请

US20130346988A1 PARALLEL DATA COMPUTING OPTIMIZATION 有权
标题翻译：并行数据计算优化
公开(公告)号：US20130346988A1
公开(公告)日：2013-12-26
申请号：US13530490
申请日：2012-06-22
申请人： Nicolas Bruno , Jingren Zhou , Srikanth Kandula , Sameer Agarwal , Ming-Chuan Wu
发明人： Nicolas Bruno , Jingren Zhou , Srikanth Kandula , Sameer Agarwal , Ming-Chuan Wu
IPC分类号： G06F9/46
CPC分类号： G06F9/5066
摘要： The use of statistics collected during the parallel distributed execution of the tasks of a job may be used to optimize the performance of the task or similar recurring tasks. An execution plan for a job is initially generated, in which the execution plan includes tasks. Statistics regarding operations performed in the tasks are collected while the tasks are executed via parallel distributed execution. Another execution plan is then generated for another recurring job, in which the additional execution plan has at least one task in common with the execution plan for the job. The additional execution plan is subsequently optimized based at least on the statistics to produce an optimized execution plan.
摘要翻译：在并行分布式执行作业期间收集的统计数据的使用可用于优化任务的性能或类似的重复任务。最初生成作业的执行计划，其中执行计划包括任务。在通过并行分布式执行执行任务时，将收集在任务中执行的操作的统计信息。然后为另一个循环作业生成另一执行计划，其中附加执行计划具有至少一个与该作业的执行计划相同的任务。随后根据统计数据优化附加执行计划，以生成优化的执行计划。

2. 发明授权

US09235446B2 Parallel computing execution plan optimization 有权
标题翻译：并行计算执行计划优化
公开(公告)号：US09235446B2
公开(公告)日：2016-01-12
申请号：US13530490
申请日：2012-06-22
申请人： Nicolas Bruno , Jingren Zhou , Srikanth Kandula , Sameer Agarwal , Ming-Chuan Wu
发明人： Nicolas Bruno , Jingren Zhou , Srikanth Kandula , Sameer Agarwal , Ming-Chuan Wu
IPC分类号： G06F9/46 , G06F17/30 , G06F9/50
CPC分类号： G06F9/5066
摘要： The use of statistics collected during the parallel distributed execution of the tasks of a job may be used to optimize the performance of the task or similar recurring tasks. An execution plan for a job is initially generated, in which the execution plan includes tasks. Statistics regarding operations performed in the tasks are collected while the tasks are executed via parallel distributed execution. Another execution plan is then generated for another recurring job, in which the additional execution plan has at least one task in common with the execution plan for the job. The additional execution plan is subsequently optimized based at least on the statistics to produce an optimized execution plan.
摘要翻译：在并行分布式执行作业期间收集的统计数据的使用可用于优化任务的性能或类似的重复任务。最初生成作业的执行计划，其中执行计划包括任务。在通过并行分布式执行执行任务时，将收集在任务中执行的操作的统计信息。然后为另一个循环作业生成另一执行计划，其中附加执行计划具有至少一个与该作业的执行计划相同的任务。随后根据统计数据优化附加执行计划，以生成优化的执行计划。

3. 发明授权

US08996464B2 Efficient partitioning techniques for massively distributed computation 有权
标题翻译：用于大规模分布式计算的高效分区技术
公开(公告)号：US08996464B2
公开(公告)日：2015-03-31
申请号：US13494006
申请日：2012-06-11
申请人： Jingren Zhou , Nicolas Bruno , Wei Lin
发明人： Jingren Zhou , Nicolas Bruno , Wei Lin
IPC分类号： G06F17/30 , G06F9/50
CPC分类号： G06F9/5066 , G06F17/30584 , Y10S707/968
摘要： A repartitioning optimizer identifies alternative repartitioning strategies and selects optimal ones, accounting for network transfer utilization and partition sizes in addition to traditional metrics. If prior partitioning was hash-based, the repartitioning optimizer can determine whether a hash-based repartitioning can result in not every computing device providing data to every other computing device. If prior partitioning was range-based, the repartitioning optimizer can determine whether a range-based repartitioning can generate similarly sized output partitions while aligning input and output partition boundaries, increasing the number of computing devices that do not provide data to every other computing device. Individual computing devices, as they are performing a repartitioning, assign a repartitioning index to each individual data element, which represents the computing device to which such a data element is destined. The indexed data is sorted by such repartitioning indices, thereby grouping together all like data, and then stored in a sequential manner.
摘要翻译：重新分配优化器识别替代重新分配策略并选择最优策略，除传统度量之外还考虑网络传输利用率和分区大小。如果先前分区是基于哈希的，则重新分区优化器可以确定基于哈希的重新分区是否可以导致不是每个计算设备向每个其他计算设备提供数据。如果先前分区是基于范围的，则重新分配优化器可以确定基于范围的重新分区是否可以在对齐输入和输出分区边界的同时生成类似大小的输出分区，从而增加不向每个其他计算设备提供数据的计算设备的数量。当各个计算设备正在执行重新分区时，将重新分配索引分配给每个单独的数据元素，其表示这样的数据元素注定到的计算设备。索引数据通过这样的重新分配索引排序，从而将所有相似数据分组在一起，然后以顺序方式存储。

4. 发明申请

US20130332446A1 EFFICIENT PARTITIONING TECHNIQUES FOR MASSIVELY DISTRIBUTED COMPUTATION 有权
标题翻译：用于大规模分布式计算的有效分区技术
公开(公告)号：US20130332446A1
公开(公告)日：2013-12-12
申请号：US13494006
申请日：2012-06-11
申请人： Jingren Zhou , Nicolas Bruno , Wei Lin
发明人： Jingren Zhou , Nicolas Bruno , Wei Lin
IPC分类号： G06F17/30
CPC分类号： G06F9/5066 , G06F17/30584 , Y10S707/968
摘要： A repartitioning optimizer identifies alternative repartitioning strategies and selects optimal ones, accounting for network transfer utilization and partition sizes in addition to traditional metrics. If prior partitioning was hash-based, the repartitioning optimizer can determine whether a hash-based repartitioning can result in not every computing device providing data to every other computing device. If prior partitioning was range-based, the repartitioning optimizer can determine whether a range-based repartitioning can generate similarly sized output partitions while aligning input and output partition boundaries, increasing the number of computing devices that do not provide data to every other computing device. Individual computing devices, as they are performing a repartitioning, assign a repartitioning index to each individual data element, which represents the computing device to which such a data element is destined. The indexed data is sorted by such repartitioning indices, thereby grouping together all like data, and then stored in a sequential manner.
摘要翻译：重新分配优化器识别替代重新分配策略并选择最优策略，除传统度量之外还考虑网络传输利用率和分区大小。如果先前分区是基于哈希的，则重新分区优化器可以确定基于哈希的重新分区是否可以导致不是每个计算设备向每个其他计算设备提供数据。如果先前分区是基于范围的，则重新分配优化器可以确定基于范围的重新分区是否可以在对齐输入和输出分区边界的同时生成类似大小的输出分区，从而增加不向每个其他计算设备提供数据的计算设备的数量。当各个计算设备正在执行重新分区时，将重新分配索引分配给每个单独的数据元素，其表示这样的数据元素注定到的计算设备。索引数据通过这样的重新分配索引排序，从而将所有相似数据分组在一起，然后以顺序方式存储。

5. 发明授权

US08966486B2 Distributed multi-phase batch job processing 有权
标题翻译：分布式多阶段批处理作业处理
公开(公告)号：US08966486B2
公开(公告)日：2015-02-24
申请号：US13099814
申请日：2011-05-03
申请人： Thomas Phan , Jingren Zhou
发明人： Thomas Phan , Jingren Zhou
IPC分类号： G06F9/46 , G06F9/48
CPC分类号： G06F9/4843
摘要： A distributed job-processing environment including a server, or servers, capable of receiving and processing user-submitted job queries for data sets on backend storage servers. The server identifies computational tasks to be completed on the job as well as a time frame to complete some of the computational tasks. Computational tasks may include, without limitation, preprocessing, parsing, importing, verifying dependencies, retrieving relevant metadata, checking syntax and semantics, optimizing, compiling, and running. The server performs the computational tasks, and once the time frame expires, a message is transmitted to the user indicating which tasks have been completed. The rest of the computational tasks are subsequently performed, and eventually, job results are transmitted to the user.
摘要翻译：分布式作业处理环境，包括能够接收和处理用户提交的后端存储服务器上的数据集作业查询的服务器或服务器。服务器识别在作业上完成的计算任务以及完成一些计算任务的时间框架。计算任务可以包括但不限于预处理，解析，导入，验证相关性，检索相关元数据，检查语法和语义，优化，编译和运行。服务器执行计算任务，一旦时间段到期，向用户发送一条消息，指示哪些任务已经完成。随后执行其余的计算任务，最终将作业结果传送给用户。

6. 发明授权

US07467128B2 Maintenance of materialized outer-join views 有权
标题翻译：维护实体外连接视图
公开(公告)号：US07467128B2
公开(公告)日：2008-12-16
申请号：US11354375
申请日：2006-02-15
申请人： Per-Ake Larson , Jingren Zhou
发明人： Per-Ake Larson , Jingren Zhou
IPC分类号： G06F17/30
CPC分类号： G06F17/30442 , Y10S707/99932
摘要： Maintenance of materialized view for query optimization in a database. The architecture provides the first practical algorithms for incrementally maintaining views with multiple outer joins and aggregated outer-join views. Materialized views with any combination of inner joins, one-sided outer joins and full outer joins, plus an optional aggregation on top, can be processed by algorithm(s) that construct incremental maintenance expressions based on converting the view definition a join-disjunctive normal form and exploiting database constraints to reduce maintenance overhead. A system comprises a view creation component for creating a materialized view definition that includes outer-joins, a conversion component for converting an outer-join view expression into a normal form, and a maintenance component for updating the stored materialized view result associated with the materialized view definition.
摘要翻译：维护数据库中查询优化的物化视图。该架构提供了第一个实际算法，用于逐步维护具有多个外连接和聚合外连接视图的视图。可以通过基于通过将视图定义转换为连接分离正常来构造增量维护表达式的算法来处理具有任何内部联接，单面外部连接和完全外部连接的组合的物化视图以及顶部的可选聚合形成和利用数据库约束来减少维护开销。一种系统包括用于创建包括外连接的物化视图定义的视图创建组件，用于将外连接视图表达式转换为正常形式的转换组件，以及用于更新与实例化相关联的存储的物化视图结果的维护组件视图定义。

7. 发明授权

US08819017B2 Affinitizing datasets based on efficient query processing 有权
标题翻译：基于有效的查询处理关联数据集
公开(公告)号：US08819017B2
公开(公告)日：2014-08-26
申请号：US12905464
申请日：2010-10-15
申请人： Jingren Zhou , Patrick James Helland , Jonathan Forbes , Yaron Burd
发明人： Jingren Zhou , Patrick James Helland , Jonathan Forbes , Yaron Burd
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06F17/30982
摘要： Embodiments of the present invention relate to systems, methods, and computer-storage media for affinitizing datasets based on efficient query processing. In one embodiment, a plurality of datasets within a data stream is received. The data stream is partitioned based on efficient query processing. Once the data stream is partitioned, an affinity identifier is assigned to datasets based on the partitioning of the dataset. Further, when datasets are broken into extents, the affinity identifier of the parent dataset is retained in the resulting extent. The affinity identifier of each extent is then referenced to preferentially store extents having common affinity identifiers within close proximity of one other across a data center.
摘要翻译：本发明的实施例涉及用于基于有效查询处理关联数据集的系统，方法和计算机存储介质。在一个实施例中，接收数据流内的多个数据集。基于有效的查询处理对数据流进行分区。一旦数据流被分区，基于数据集的分区，将一个亲和度标识符分配给数据集。此外，当数据集分成多个区段时，父数据集的亲和性标识符将保留在生成的范围内。然后引用每个范围的相似性标识符，以优先地存储具有跨越数据中心的彼此靠近的共同相似性标识符的盘区。

8. 发明申请

US20090327255A1 VIEW MATCHING OF MATERIALIZED XML VIEWS 有权
标题翻译：查看材料化XML视图的匹配
公开(公告)号：US20090327255A1
公开(公告)日：2009-12-31
申请号：US12146506
申请日：2008-06-26
申请人： Per-Ake Larson , Guido Moerkotte , Frank W. Tompa , Jingren Zhou
发明人： Per-Ake Larson , Guido Moerkotte , Frank W. Tompa , Jingren Zhou
IPC分类号： G06F7/06 , G06F17/30
CPC分类号： G06F17/30457 , G06F17/30935
摘要： A materialized XML view matching system and method for processing of SQLXML queries using view matching of materialized XML views. The view matching process of the embodiments of the system and method use a multi-path tree (MPT) data structure. Embodiments of the materialized XML view matching system and method construct an MPT data structure for each input query and view expression. View matching is performed on the MPT data structures to generate a set of partial matches, which then are cleaned to generate a set of candidate matches. A valid match definition is generated by testing each candidate match for different forms of compliance. Using the valid match definition, a set of valid matches is identified and extracted. For each valid match, a substitute query expression is constructed that can serve as a replacement for the original query. These substitute queries can be used to evaluate the original query.
摘要翻译：一种物化XML视图匹配系统和方法，用于使用实例化XML视图的视图匹配来处理SQLXML查询。系统和方法的实施例的视图匹配过程使用多路径树（MPT）数据结构。物化XML视图匹配系统和方法的实施例为每个输入查询和视图表达构造一个MPT数据结构。在MPT数据结构上执行视图匹配以生成一组部分匹配，然后对其进行清理以生成一组候选匹配。通过测试不同形式的合规性的每个候选匹配来生成有效的匹配定义。使用有效的匹配定义，识别并提取一组有效的匹配。对于每个有效的匹配，构造可替代原始查询的替代查询表达式。这些替代查询可用于评估原始查询。

9. 发明授权

US08140495B2 Asynchronous database index maintenance 有权
标题翻译：异步数据库索引维护
公开(公告)号：US08140495B2
公开(公告)日：2012-03-20
申请号：US12435026
申请日：2009-05-04
申请人： Peter A. Carlin , Per-Ake Larson , Jingren Zhou
发明人： Peter A. Carlin , Per-Ake Larson , Jingren Zhou
IPC分类号： G06F17/30
CPC分类号： G06F17/30312
摘要： This disclosure provides techniques for asynchronously maintaining database indexes or sub-indexes. For example, a database management server may receive a data manipulation statement to modify particular data stored in a database and determine whether an index associated with executing the statement is maintained asynchronously. When the index is maintained asynchronously, maintenance of the index to reflect changes made to the particular data by executing the data manipulation statement may be delayed until an index maintenance event. The index maintenance may be based on an isolation level of a transaction including a query that triggered the index maintenance.
摘要翻译：本公开提供用于异步维护数据库索引或子索引的技术。例如，数据库管理服务器可以接收数据操作语句来修改存储在数据库中的特定数据，并且确定与执行语句相关联的索引是否被异步地维护。当索引异步维护时，通过执行数据操作语句来反映对特定数据所做的更改的索引的维护可能会延迟到索引维护事件。索引维护可以基于事务的隔离级别，包括触发索引维护的查询。

10. 发明授权

US07599925B2 Using query expression signatures in view matching 失效
标题翻译：在视图匹配中使用查询表达式签名
公开(公告)号：US07599925B2
公开(公告)日：2009-10-06
申请号：US11094924
申请日：2005-03-31
申请人： Per-Ake Larson , David Edward DeHann , Jingren Zhou
发明人： Per-Ake Larson , David Edward DeHann , Jingren Zhou
IPC分类号： G06F17/30
CPC分类号： G06F17/30457 , Y10S707/99932 , Y10S707/99933 , Y10S707/99934
摘要： A system for optimizing queries against a database is described. The system comprises a memo structure that encodes a plurality of query expressions. The system also includes a signature mechanism that enables an assignment of the query expressions into equivalence classes. Methods of using such a system are additionally provided.
摘要翻译：描述用于针对数据库优化查询的系统。该系统包括对多个查询表达式进行编码的备忘录结构。该系统还包括一个签名机制，可以将查询表达式分配为等价类。另外提供使用这种系统的方法。

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