公开(公告)号：US20140280020A1

公开(公告)日：2014-09-18

申请号：US13802023

申请日：2013-03-13

Applicant: FUTUREWEI TECHNOLOGIES, INC.

Inventor： Mohan Singamshetty ,  Yang Sun ,  Qingqing Zhou

IPC: G06F17/30

CPC classification number: G06F17/30445

Abstract: System and method embodiments are provided for improving the performance of query processing in a massively parallel processing (MPP) database system by pushing down join query processing to data nodes recursively. An embodiment method includes receiving, at a coordinator process, a join query associated with a plurality of tables of the MPP database system, generating, at the coordinator process, an execution plan tree for the join query, and processing, at each of a plurality of data nodes communicating with the coordinator process, the execution plan tree to obtain join query results. The method further includes, upon detecting a next join operator below a top join operator in the execution plan tree at each of the data nodes, forwarding to the other data nodes a sub-tree for the next join operator, and receiving, at each of the data nodes from the other data nodes, sub-tree processing results.

Abstract translation: 提供了系统和方法实施例，用于通过将连接查询处理递归地推送到数据节点来提高大规模并行处理（MPP）数据库系统中查询处理的性能。 一种实施方式包括在协调器进程处接收与MPP数据库系统的多个表相关联的连接查询，在协调器进程处生成用于连接查询的执行计划树，并在多个表中的每一个处处理 的数据节点与协调器进程通信，执行计划树获取连接查询结果。 该方法还包括：在每个数据节点处检测执行计划树中的顶部连接运算符下面的下一个连接运算符，向下一个连接运算符转发另一个数据节点的子树，并在每个 来自其他数据节点的数据节点，子树处理结果。

公开(公告)号：US09152669B2

公开(公告)日：2015-10-06

申请号：US13802084

申请日：2013-03-13

Applicant: FutureWei Technologies, Inc.

Inventor： Mohan Singamshetty

IPC: G06F17/00 ,  G06F17/30 ,  G06F1/035

CPC classification number: G06F17/30442 ,  G06F1/035 ,  G06F17/30498

Abstract: System and method embodiments are provided for improving the performance of query processing in a massively parallel processing (MPP) database system by selecting one or more stationary tables for processing the query plan. In an embodiment method, a join query associated with a plurality of tables of the MPP database system is received at a coordinator process. Stationary tables are then selected from the tables according to table partition keys in the join query. The stationary tables are indicated and the join query are sent to a plurality of data nodes communicating with the coordinator process. At each of the data nodes, an execution plan for the join query is generated. Each of the data nodes then executes the execution plan to obtain query results without sharing information of the stationary tables between the data nodes.

Abstract translation: 提供了系统和方法实施例，用于通过选择用于处理查询计划的一个或多个固定表来改善大规模并行处理（MPP）数据库系统中的查询处理的性能。 在一个实施方式中，在协调器进程处接收与MPP数据库系统的多个表相关联的连接查询。 然后根据连接查询中的表分区键从表中选择固定表。 指示固定表，并将连接查询发送到与协调器进程通信的多个数据节点。 在每个数据节点上，生成连接查询的执行计划。 然后，每个数据节点执行执行计划以获得查询结果，而不共享数据节点之间的静态表的信息。

公开(公告)号：US20140188841A1

公开(公告)日：2014-07-03

申请号：US13730872

申请日：2012-12-29

Applicant: FUTUREWEI TECHNOLOGIES, INC.

Inventor： Yang Sun ,  Qingqing Zhou ,  Mohan Singamshetty

IPC: G06F17/30

CPC classification number: G06F17/30445 ,  G06F17/30483

Abstract: Queries may be processed more efficiently in an massively parallel processing (MPP) database by locally optimizing the global execution plan. The global execution plan and a semantic tree may be provided to MPP data nodes by an MPP coordinator. The MPP data nodes may then use the global execution plan and the semantic tree to generate a local execution plan. Thereafter, the MPP data nodes may select either the global execution plan or the local execution plan is accordance with a cost evaluation.

Abstract translation: 在大规模并行处理（MPP）数据库中，可以通过局部优化全局执行计划，更有效地处理查询。 全局执行计划和语义树可以由MPP协调器提供给MPP数据节点。 然后，MPP数据节点可以使用全局执行计划和语义树来生成本地执行计划。 此后，MPP数据节点可以选择全局执行计划，或者本地执行计划根据成本评估。

公开(公告)号：US09576026B2

公开(公告)日：2017-02-21

申请号：US13802023

申请日：2013-03-13

Applicant: FutureWei Technologies, Inc.

Inventor： Mohan Singamshetty ,  Yang Sun ,  Qingqing Zhou

IPC: G06F17/30

CPC classification number: G06F17/30445

Abstract: System and method embodiments are provided for improving the performance of query processing in a massively parallel processing (MPP) database system by pushing down join query processing to data nodes recursively. An embodiment method includes receiving, at a coordinator process, a join query associated with a plurality of tables of the MPP database system, generating, at the coordinator process, an execution plan tree for the join query, and processing, at each of a plurality of data nodes communicating with the coordinator process, the execution plan tree to obtain join query results. The method further includes, upon detecting a next join operator below a top join operator in the execution plan tree at each of the data nodes, forwarding to the other data nodes a sub-tree for the next join operator, and receiving, at each of the data nodes from the other data nodes, sub-tree processing results.

Abstract translation: 提供了系统和方法实施例，用于通过将连接查询处理递归地推送到数据节点来提高大规模并行处理（MPP）数据库系统中查询处理的性能。 一种实施方式包括在协调器进程处接收与MPP数据库系统的多个表相关联的连接查询，在协调器进程处生成用于连接查询的执行计划树，并在多个表中的每一个处处理 的数据节点与协调器进程通信，执行计划树获取连接查询结果。 该方法还包括：在每个数据节点处检测执行计划树中的顶部连接运算符下面的下一个连接运算符，向下一个连接运算符转发另一个数据节点的子树，并在每个 来自其他数据节点的数据节点，子树处理结果。

公开(公告)号：US09311354B2

公开(公告)日：2016-04-12

申请号：US13730872

申请日：2012-12-29

Applicant: FutureWei Technologies, Inc.

Inventor： Jason Yang Sun ,  Qingqing Zhou ,  Mohan Singamshetty

IPC: G06F17/30

CPC classification number: G06F17/30445 ,  G06F17/30483

Abstract: Queries may be processed more efficiently in an massively parallel processing (MPP) database by locally optimizing the global execution plan. The global execution plan and a semantic tree may be provided to MPP data nodes by an MPP coordinator. The MPP data nodes may then use the global execution plan and the semantic tree to generate a local execution plan. Thereafter, the MPP data nodes may select either the global execution plan or the local execution plan is accordance with a cost evaluation.

Abstract translation: 在大规模并行处理（MPP）数据库中，可以通过局部优化全局执行计划，更有效地处理查询。 全局执行计划和语义树可以由MPP协调器提供给MPP数据节点。 然后，MPP数据节点可以使用全局执行计划和语义树来生成本地执行计划。 此后，MPP数据节点可以选择全局执行计划，或者本地执行计划根据成本评估。

公开(公告)号：US20140280021A1

公开(公告)日：2014-09-18

申请号：US13802084

申请日：2013-03-13

Applicant: FUTUREWEI TECHNOLOGIES, INC.

Inventor： Mohan Singamshetty

IPC: G06F17/30

CPC classification number: G06F17/30442 ,  G06F1/035 ,  G06F17/30498

Abstract: System and method embodiments are provided for improving the performance of query processing in a massively parallel processing (MPP) database system by selecting one or more stationary tables for processing the query plan. In an embodiment method, a join query associated with a plurality of tables of the MPP database system is received at a coordinator process. Stationary tables are then selected from the tables according to table partition keys in the join query. The stationary tables are indicated and the join query are sent to a plurality of data nodes communicating with the coordinator process. At each of the data nodes, an execution plan for the join query is generated. Each of the data nodes then executes the execution plan to obtain query results without sharing information of the stationary tables between the data nodes.

Abstract translation: 提供了系统和方法实施例，用于通过选择用于处理查询计划的一个或多个固定表来改善大规模并行处理（MPP）数据库系统中的查询处理的性能。 在一个实施方式中，在协调器进程处接收与MPP数据库系统的多个表相关联的连接查询。 然后根据连接查询中的表分区键从表中选择固定表。 指示固定表，并将连接查询发送到与协调器进程通信的多个数据节点。 在每个数据节点上，生成连接查询的执行计划。 然后，每个数据节点执行执行计划以获得查询结果，而不共享数据节点之间的静态表的信息。