公开(公告)号：US20240095260A1

公开(公告)日：2024-03-21

申请号：US18520127

申请日：2023-11-27

Applicant: HUAWEI TECHNOLOGIES CO., LTD. ,  Fudan University

Inventor： Zhen Tian ,  Weiguo Zheng ,  Junhua Zhu

IPC: G06F16/28

CPC classification number: G06F16/287

Abstract: A multi-subgraph matching method and apparatus, and a device are provided. After receiving a plurality of query graphs, the multi-subgraph matching apparatus groups the plurality of query graphs based on a hash value of each query graph, to generate a plurality of groups of query graphs. A plurality of query graphs whose hash values fall within a same value range belong to a same group. Then, the multi-subgraph matching apparatus respectively matches the plurality of groups of query graphs with a data graph in parallel, to obtain matching results. The matching results are matching results between the plurality of query graphs and the data graph. According to the multi-subgraph matching method in this application, grouping efficiency can be increased, and subgraph matching efficiency can be effectively increased.

公开(公告)号：US20190324952A1

公开(公告)日：2019-10-24

申请号：US16458684

申请日：2019-07-01

Applicant: Huawei Technologies Co., Ltd.

Inventor： Shanshan Ying ,  Junhua Zhu

IPC: G06F16/21 ,  G06F16/242 ,  G06F16/2455

Abstract: A stored-procedure execution method includes receiving an execution request, where the execution request is used to request to execute a task including at least two stored procedures, requesting at least two threads, and dispatching each stored procedure in the task to one of the at least two threads for execution, receiving Structured Query Language (SQL) statements sent by the at least two threads when the at least two threads execute the stored procedures included in the task, and grouping and caching the received SQL statements based on a same access characteristic, and for an SQL statement cache group that satisfies a preset trigger condition, calling an SQL statement execution engine to execute an SQL statement in the SQL statement cache group.

公开(公告)号：US10359939B2

公开(公告)日：2019-07-23

申请号：US14801421

申请日：2015-07-16

Applicant: Huawei Technologies Co., Ltd.

Inventor： Jiansheng Wei ,  Junhua Zhu

IPC: G06F12/00 ,  G06F3/06 ,  G06F12/02

Abstract: Embodiments of the present invention provide a data object processing method and apparatus, which can divide a data object into one or more blocks; calculate a sample compression ratio of each block, aggregate neighboring consecutive blocks with a same sample compression ratio characteristic into one data segment, and obtain the sample compression ratio of each of the data segments; and select, according to a length range to which a length of each of the data segments belongs and a compression ratio range to which the sample compression ratio of each of the data segments belongs, an expected length to divide the data segment into data chunks, where the sample compression ratio of each of the data segments uniquely belongs to one of the compression ratio ranges, and the length of each of the data segments uniquely belongs to one of the length ranges.

公开(公告)号：US20180276277A1

公开(公告)日：2018-09-27

申请号：US15992925

申请日：2018-05-30

Applicant: Huawei Technologies Co., Ltd.

Inventor： Guoping Wang ,  Junhua Zhu

IPC: G06F17/30

CPC classification number: G06F16/24544 ,  G06F16/24545

Abstract: A data query method and apparatus, and a database system, where the method includes receiving a data query request, generating an original query plan according to the data query request, obtaining a candidate query plan set according to the original query plan, restructuring a join predicate in the original query plan, determining basic information of a restructured join predicate in the original query plan, determining a constraint condition of the restructured join predicate in the original query plan, determining an equal-cost query plan for the original query plan according to the basic information of the restructured join predicate in the original query plan and the constraint condition of the restructured join predicate in the original query plan, and performing querying according to the equal-cost query plan for the original query plan. Hence, data query performance can be improved.

公开(公告)号：US20160357740A1

公开(公告)日：2016-12-08

申请号：US15241909

申请日：2016-08-19

Applicant: Huawei Technologies Co., Ltd.

Inventor： Liqun Deng ,  Junhua Zhu

IPC: G06F17/30

CPC classification number: G06F17/3012 ,  G06F3/061 ,  G06F3/0638 ,  G06F3/067 ,  G06F17/30091 ,  G06F17/30185 ,  G06F17/30194 ,  G06F17/30238

Abstract: A metadata updating method based on columnar storage in a distributed file system includes acquiring to-be-updated metadata in a data table, splitting data records of the data table into multiple row groups on a row basis, converting the data table into global file metadata and multiple row group files, where the row group file includes an actual data block, a data index block, a local metadata block, a metadata index block, and a file footer, determining whether the to-be-updated metadata belongs to the global file metadata, updating local metadata when the to-be-updated metadata does not belong to the global file metadata, and adding an updated local metadata block, an updated metadata index block, and an updated file footer to the multiple row group files according to updated local metadata. Dynamic updating of metadata saves time of executing an updating operation of this type and needed computing resources.

Abstract translation: 基于分布式文件系统中的柱状存储的元数据更新方法包括获取数据表中的要更新的元数据，将数据表的数据记录划分为行的多个行组，将数据表转换为全局文件元数据 以及多个行组文件，其中行组文件包括实际数据块，数据索引块，本地元数据块，元数据索引块和文件页脚，确定要更新的元数据是否属于全局 文件元数据，当待更新的元数据不属于全局文件元数据时，更新本地元数据，并且根据更新的本地元数据块，更新的元数据索引块和更新的文件页脚添加到多个行组文件 更新的本地元数据。 元数据的动态更新节省了执行此类更新操作的时间，并需要计算资源。

公开(公告)号：US20150278030A1

公开(公告)日：2015-10-01

申请号：US14738062

申请日：2015-06-12

Applicant: Huawei Technologies Co., Ltd.

Inventor： Junhua Zhu

IPC: G06F11/14 ,  G06F17/30

CPC classification number: G06F11/1451 ,  G06F11/1464 ,  G06F11/2038 ,  G06F11/2048 ,  G06F11/2097 ,  G06F16/2255 ,  G06F16/273 ,  G06F16/275 ,  G06F2201/80 ,  G06F2201/82 ,  G06F2201/84

Abstract: A distributed database synchronization method and system. A distributed database includes a master server cluster and a backup server cluster, where the master server cluster includes a first master node and a second master node, and the backup server cluster includes a first backup node and a second backup node. The method includes: generating a hash tree of the master server cluster and a hash tree of the backup server cluster; determining a range hash tree of the second master node and a range hash tree of the second backup node that have inconsistent range hash values; determining a data unit to be synchronized in the second master node and a data unit to be synchronized in the second backup node; and performing data synchronization. Because data units to be synchronized are determined separately and simultaneously in multiple nodes, thereby improving efficiency of data synchronization.

Abstract translation: 分布式数据库同步方法和系统。 分布式数据库包括主服务器集群和备份服务器集群，其中主服务器集群包括第一主节点和第二主节点，备份服务器集群包括第一备份节点和第二备份节点。 该方法包括：生成主服务器集群的散列树和备份服务器集群的散列树; 确定第二主节点的范围散列树和具有不一致的范围散列值的第二备份节点的范围散列树; 确定要在所述第二主节点中同步的数据单元和要在所述第二备份节点中同步的数据单元; 并执行数据同步。 因为要同步的数据单元在多个节点中单独并同时确定，从而提高数据同步的效率。

公开(公告)号：US20150074345A1

公开(公告)日：2015-03-12

申请号：US14546797

申请日：2014-11-18

Applicant: Huawei Technologies Co., Ltd.

Inventor： Junhua Zhu

IPC: G06F12/08 ,  G06F12/02

CPC classification number: G06F12/123 ,  G06F12/0246 ,  G06F12/0871 ,  G06F12/121 ,  G06F2212/214 ,  G06F2212/222 ,  G06F2212/502 ,  G06F2212/604 ,  G06F2212/7204

Abstract: Embodiments of the present invention disclose a method and apparatus of cache management for a non-volatile storage device. The method embodiment includes: determining a size relationship between a capacity sum of a clean page subpool and a dirty page subpool and a cache capacity; determining, when the capacity sum is equal to the cache capacity, whether identification information of a to-be-accessed page is in a history list of clean pages or a history list of dirty pages; and when it is determined that the identification information of the to-be-accessed page is in the history list of clean pages, adding a first adjustment value to a clean page subpool capacity threshold; and when the identification information of the to-be-accessed page is in the history list of dirty pages, subtracting a second adjustment value from the clean page subpool capacity threshold.

Abstract translation: 本发明的实施例公开了一种用于非易失性存储设备的高速缓存管理的方法和装置。 方法实施例包括：确定清洁页面子站和脏页面子空间的容量和与高速缓存容量之间的大小关系; 当容量和等于高速缓存容量时，确定待访问页面的识别信息是否在干净页面的历史列表或脏页面的历史列表中; 并且当确定要访问页面的识别信息在清洁页面的历史列表中时，将第一调整值添加到清洁页面子库容量阈值; 并且当待访问页面的识别信息在脏页面的历史列表中时，从清洁页面子库容量阈值减去第二调整值。

公开(公告)号：US10885034B2

公开(公告)日：2021-01-05

申请号：US15992925

申请日：2018-05-30

Applicant: Huawei Technologies Co., Ltd.

Inventor： Guoping Wang ,  Junhua Zhu

IPC: G06F16/2453

Abstract: A data query method and apparatus, and a database system, where the method includes receiving a data query request, generating an original query plan according to the data query request, obtaining a candidate query plan set according to the original query plan, restructuring a join predicate in the original query plan, determining basic information of a restructured join predicate in the original query plan, determining a constraint condition of the restructured join predicate in the original query plan, determining an equal-cost query plan for the original query plan according to the basic information of the restructured join predicate in the original query plan and the constraint condition of the restructured join predicate in the original query plan, and performing querying according to the equal-cost query plan for the original query plan. Hence, data query performance can be improved.

公开(公告)号：US10558498B2

公开(公告)日：2020-02-11

申请号：US15598696

申请日：2017-05-18

Applicant: Huawei Technologies Co., Ltd.

Inventor： Liqun Deng ,  Junhua Zhu

IPC: G06F9/46 ,  G06F9/50 ,  G06F9/48

Abstract: A method for scheduling a data flow task and an apparatus. The method includes: preprocessing a data flow task to obtain at least one subtask; classifying the subtask into a central processing unit (CPU) task group, a graphics processing unit (GPU) task group, or a to-be-determined task group; allocating the subtask to a working node; when the subtask belongs to the CPU task group, determining that a CPU executes the subtask; when the subtask belongs to the GPU task group, determining that a GPU executes the subtask; or when the subtask belongs to the to-be-determined task group, determining, according to costs of executing the subtask by a CPU and a GPU, a running platform (e.g., the CPU or the GPU) executes the subtask, where the cost includes duration of executing the subtask.

公开(公告)号：US20170255496A1

公开(公告)日：2017-09-07

申请号：US15598696

申请日：2017-05-18

Applicant: Huawei Technologies Co., Ltd.

Inventor： Liqun Deng ,  Junhua Zhu

IPC: G06F9/50 ,  G06F9/48

CPC classification number: G06F9/5027 ,  G06F9/4881 ,  G06F9/5044

Abstract: A method for scheduling a data flow task and an apparatus. The method includes: preprocessing a data flow task to obtain at least one subtask; classifying the subtask into a central processing unit (CPU) task group, a graphics processing unit (GPU) task group, or a to-be-determined task group; allocating the subtask to a working node; when the subtask belongs to the CPU task group, determining that a CPU executes the subtask; when the subtask belongs to the GPU task group, determining that a GPU executes the subtask; or when the subtask belongs to the to-be-determined task group, determining, according to costs of executing the subtask by a CPU and a GPU, a running platform (e.g., the CPU or the GPU) executes the subtask, where the cost includes duration of executing the subtask.