WO2012065478A1 - 一种智能推送数据的方法和装置 - Google Patents

一种智能推送数据的方法和装置 Download PDF

Info

Publication number
WO2012065478A1
WO2012065478A1 PCT/CN2011/080098 CN2011080098W WO2012065478A1 WO 2012065478 A1 WO2012065478 A1 WO 2012065478A1 CN 2011080098 W CN2011080098 W CN 2011080098W WO 2012065478 A1 WO2012065478 A1 WO 2012065478A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
user
push
specified
pushing
Prior art date
Application number
PCT/CN2011/080098
Other languages
English (en)
French (fr)
Inventor
蔡桂英
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2012065478A1 publication Critical patent/WO2012065478A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/21Design, administration or maintenance of databases
    • G06F16/217Database tuning
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/27Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor

Definitions

  • the present invention relates to a method and apparatus for intelligently pushing data to actively pre-push data on a distributed database data source side.
  • the use of data and the data source are physically separated.
  • the application must obtain data from the data source before processing.
  • a large number of data sets from the data source to the application not only consume network bandwidth, but also increase
  • the application waits and processes time, which in turn affects the application's response time.
  • cached data are all "old" data, that is, the used unupdated may be inconsistent with the source data, when the customer
  • the response time is significantly lower than the average response time.
  • the load of the network and the distributed database is also increased instantaneously. When the client needs a large amount of new data, the response speed of the client still has a bottleneck.
  • the technical problem to be solved by the present invention is to provide a method and device for intelligently pushing data, so as to actively push data to the client, and greatly improve the response speed of the client.
  • the present invention provides a method for intelligently pushing data, including: acquiring a push mechanism related to a user; and determining pre-push data according to the push mechanism;
  • the pushing mechanism includes:
  • a field set mechanism for pushing data in a user-specified field set includes:
  • the field set mechanism for pushing data in the field set specified by the user is selected.
  • the above method further has the following features: the step of determining the pre-push data according to the pushing mechanism comprises: determining the pre-push data according to the pushing mechanism and a historical data requirement of the user.
  • the foregoing method further has the following features: the step of determining the pre-push data according to the pushing mechanism further includes:
  • the user's historical data requirements are updated.
  • the foregoing method further has the following features: after the step of receiving the data requirement of the user, the method further includes:
  • the field accessed by the user at the time is added to the specified field set;
  • the table accessed by the user at that time is added to the specified table set.
  • the foregoing method further has the following features: after the step of pushing the pre-push data to the specified cache, the method further includes:
  • the historical data requirement of the user is analyzed, and the push mechanism is re-determined.
  • the foregoing method further has the following features: the analyzing the historical data requirement of the user, and the step of re-determining the pushing mechanism includes:
  • the data requirements of the users who miss the target data for the first time are analyzed one by one, and the pre-push data is verified according to the corresponding push mechanism assumption, and the hypothesis pre-push data is verified.
  • the hit ratio if the hit ratio is greater than a specified value, determines the corresponding push mechanism as a push mechanism associated with the user.
  • the present invention also provides an apparatus for intelligently pushing data, including:
  • the obtaining module is set to obtain a push mechanism related to the user;
  • a determining module configured to determine pre-push data according to the pushing mechanism
  • a push module configured to push the pre-push data to a specified cache.
  • the pushing mechanism includes:
  • a field set mechanism for pushing data in a user-specified field set a table-like mechanism for pushing data in a table accessed by a user, a table collection mechanism for pushing data of a table in a user-specified table set, a push recently used, and being updated
  • the latest mechanism for data or an automatic mechanism for automatically determining the priority of pushed data The latest mechanism for data or an automatic mechanism for automatically determining the priority of pushed data.
  • the acquiring module is specifically configured to: if it is determined that the user is the first login, obtain a field set mechanism for pushing data in a field set specified by the user.
  • the determining module is specifically configured to determine the pre-push data according to the pushing mechanism and the historical data requirement of the user.
  • the above device further has the following features: further comprising: an update module configured to receive the user's data requirements, and update the user's historical data requirements.
  • the foregoing apparatus further has the following features: the updating module is specifically configured to: if detecting that the data requirement does not belong to a field in the specified field set, add a field accessed by the user to the specified field set; or If it is detected that the data requirement does not belong to a field in the specified table set, the table accessed by the user at the time is added to the specified table set.
  • the above device further has the following features:
  • the analysis module is configured to analyze the historical data requirement of the user and determine the push mechanism if it is determined that the number of consecutive missed target data in the cache is greater than the specified number of times according to the data requirement of the user.
  • the above device also has the following features:
  • the analyzing module is specifically configured to: if it is determined that the number of times the target data is continuously missed in the cache according to the data requirement of the user is greater than the specified number of times, sequentially adopting a pushing mechanism other than the currently used pushing mechanism, from the first
  • the data requirements of the users who missed the target data begin to analyze one by one, and pre-push the data according to the corresponding push mechanism hypothesis, and test the hit rate of the hypothetical pre-push data, if If the hit ratio is greater than the specified value, the corresponding push mechanism is determined to be a push mechanism associated with the user.
  • the above apparatus further has the following features: further comprising: a configuration module configured to accept a push mechanism configured by the user.
  • the foregoing technical solution provides a method and device for intelligently pushing data, which can actively predict and push data required by the client, can greatly reduce the data record of the client miss, and greatly improve the average response time of the client. And can make the database and network load more balanced.
  • FIG. 1 is a schematic diagram of an apparatus for intelligently pushing data according to an embodiment of the present invention
  • FIG. 2 is a flowchart of a method for intelligently pushing data according to an embodiment of the present invention. Preferred embodiment of the invention
  • the device mainly includes: a selection module, a determination module, and a push module, where
  • the obtaining module is set to obtain a push mechanism related to the user
  • a determining module configured to determine pre-push data according to the pushing mechanism
  • a push module configured to push the pre-push data to a specified cache.
  • the specified cache is preferably a cache in the client of the user.
  • the push mechanism may include: a field set mechanism for pushing data in a field set specified by a user, a same table mechanism for pushing data in a table accessed by a user, and a table set mechanism for pushing data of a table in a user specified table set. , the latest mechanism for pushing recently used and updated data, an automatic mechanism for automatically determining the data that is preferentially pushed, and so on.
  • the device of this embodiment further includes a configuration module, configured to configure a push mechanism for the corresponding user,
  • the user can be configured according to the user group, that is, the user of the same user group configures the same push mechanism.
  • the acquiring module is specifically configured to: if it is determined that the user is the first login, obtain a field set mechanism for pushing data in a field set specified by the user, if the user's field set is empty, no processing, waiting User's command.
  • the determining module is specifically configured to determine the pre-push data according to the pushing mechanism and the historical data requirement of the user.
  • the apparatus of this embodiment may further include an update module configured to update the historical data requirement of the user to receive the data requirement of the user.
  • the update module detects that the data requirement does not belong to a field in the specified field set, the field that the user accesses is added to the specified field set; or if the data requirement is not included in the specified table set.
  • the field adds the table accessed by the user to the specified table set.
  • the apparatus of this embodiment may further include: an analyzing module, configured to: if it is determined that the number of consecutive missed target data in the cache according to the data requirement of the user is greater than a specified number of times (for example, 10 times, the specified number of times is configurable), the user is The historical data needs are analyzed to re-determine the push mechanism.
  • an analyzing module configured to: if it is determined that the number of consecutive missed target data in the cache according to the data requirement of the user is greater than a specified number of times (for example, 10 times, the specified number of times is configurable), the user is The historical data needs are analyzed to re-determine the push mechanism.
  • the analyzing module determines that the number of consecutive missed target data in the cache is greater than the specified number of times according to the data requirement of the user, sequentially adopting a pushing mechanism other than the currently used pushing mechanism, from the first miss
  • the data requirements of the users of the target data are analyzed one by one, and the pre-push data is assumed according to the corresponding push mechanism, and the hit rate of the hypothetical pre-push data is checked. If the hit rate is greater than a specified value (for example, 60%), the The corresponding push mechanism is determined to be a push mechanism associated with the user.
  • the analysis module can continue to pre-analyze the user's data demand, analyze the hit rate, and when the hit rate reaches the specified value, use the corresponding The push mechanism resumes pre-pushing the data for the user.
  • FIG. 2 is a flowchart of a method for intelligently pushing data according to an embodiment of the present invention. As shown in FIG. 2, the method includes the following steps:
  • the push mechanism is a strategy for the distributed database to select the data that is actively pushed to the user.
  • the push mechanism can include a field set mechanism, a same table mechanism, a table set mechanism, a latest mechanism, an automatic mechanism, and the like.
  • the field set mechanism indicates that the data in the field set specified by the user is preferentially pushed
  • the same table mechanism indicates that other data in the table accessed by the user is preferentially pushed
  • the table collection mechanism indicates that the data of the table in the specified table set is preferentially pushed by the user
  • the latest mechanism indicates that data that has been recently used and updated has been prioritized
  • the automatic mechanism means that the data that is preferentially pushed (multi-mechanism hybrid) is automatically determined.
  • the configuration of the push mechanism can be configured according to the user group. Users of the same user group can be configured to use the same push mechanism. When the same user belongs to different user groups, the push of the first user group configured by the user can be used by default. Mechanisms, you can also specify a push mechanism when a user joins a multi-user group.
  • the distributed database side can obtain the user-related push mechanism according to the configuration information, initialize the system, and determine the pre-push data according to the user's historical data requirements, and push it to the user client.
  • the historical data requirement of the user is the data requirement when the user logs in to the distributed database last time.
  • the distributed system records and serializes the data requirement, and is re-enabled after the user logs in, which greatly reduces the response time.
  • the field set mechanism When the user logs in to the system for the first time, the field set mechanism is used to push the data in the field set. If the user's field set is empty, no processing is performed, and the user is instructed to wait.
  • the user passes the data request.
  • the client of the distributed database first checks the cache. If the target data is hit, the result is directly returned, and the user data request is sent to the distributed database side at the same time, and the distributed database side is updated.
  • the user's historical data needs, and according to the push mechanism, the user pre-push data is determined, and the data is actively pushed to the user's client.
  • the data request is sent to the distributed database side, the distributed database side returns the required data set, and the user's historical data demand record is updated, and the miss is triggered.
  • Data demand events and analyze the data requirements and The relationship between historical data needs.
  • the field accessed by the user may be added to the specified field set, where the number of fields in the field set is configurable.
  • the distributed database side detects that the data request does not belong to a field in the specified table set, the table that the user accesses may be added to the specified table set, where the number of tables in the table set is configurable.
  • the distributed database side determines that the number of consecutive missed target data in the cache according to the data requirement of the user is greater than a specified number of times (eg, 10 times, configurable), the historical data requirement needs to be analyzed, and the push mechanism is re-determined.
  • a specified number of times eg, 10 times, configurable
  • the above technical solution provides a method and device for intelligently pushing data, and actively pushing data to the client, which can greatly reduce the data record missed by the client, greatly improve the average response time of the client, improve the response speed of the client, and can The load on the database and the network is more balanced.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Information Transfer Between Computers (AREA)

Description

一种智能推送数据的方法和装置
技术领域
本发明涉及一种智能推送数据的方法和装置, 以实现在分布式数据库数 据源侧主动预推数据。
背景技术
分布式数据库中, 数据的使用和数据源在物理上是分离的, 应用程序必 须从数据源获取数据后再进行处理, 大量的数据集从数据源到应用程序不仅 耗用了网络带宽, 而且增加了应用程序的等待和处理时间, 进而影响了应用 程序的响应时间。 目前, 有很多使用在客户端緩存数据进而提高客户端响应 时间的方法, 但是这些緩存数据均为 "旧" 数据, 即使用过的未被更新过的 可能和源数据存在不一致的情况, 当客户端进行大量的新数据查询时, 响应 时间明显低于平均响应时间, 网络和分布式数据库的负载也在瞬间加大, 当 客户需要大量新的数据时, 客户端的响应速度依然存在瓶颈。
发明内容
本发明所要解决的技术问题是提供一种智能推送数据的方法和装置, 以 主动向客户端推送数据, 大大提高客户端的响应速度。 为了解决上述技术问题, 本发明提供了一种智能推送数据的方法, 包括: 获取与用户相关的推送机制; 根据所述推送机制确定出预推数据;
将所述预推数据推送到指定的緩存。
优选地, 上述方法还具有下面特点: 所述推送机制包括:
推送用户指定的字段集内的数据的字段集机制、 推送用户访问过的表内 的数据的同表机制、 推送用户指定表集合内表的数据的表集合机制、 推送最 近使用过并被更新过的数据的最新机制或自动确定优先推送的数据的自动机 制。 优选地, 上述方法还具有下面特点: 所述选取与用户相关的推送机制的 步骤包括:
若判断所述用户是第一次登录, 则选取推送用户指定的字段集内的数据 的字段集机制。
优选地, 上述方法还具有下面特点: 所述根据所述推送机制确定出预推 数据的步骤包括: 根据所述推送机制和用户的历史数据需求确定出所述预推 数据。
优选地, 上述方法还具有下面特点: 所述根据所述推送机制确定出预推 数据的步骤之前还包括:
若接收到用户的数据需求, 则更新用户的历史数据需求。
优选地, 上述方法还具有下面特点: 所述接收到用户的数据需求的步骤 之后, 还包括:
若检测所述数据需求不属于指定字段集内的字段, 则将用户该次访问的 字段增加到所述指定字段集; 或
若检测所述数据需求不属于指定表集合内的字段, 则将用户该次访问的 表增加到所述指定表集合。
优选地, 上述方法还具有下面特点: 所述将所述预推数据推送到指定的 緩存的步骤之后, 还包括:
若判断根据用户的数据需求在所述緩存中连续未命中目标数据的次数大 于指定次数, 则对用户的历史数据需求进行分析, 重新确定推送机制。
优选地, 上述方法还具有下面特点: 所述对用户的历史数据需求进行分 析, 重新确定推送机制的步骤包括:
依次釆用除当前使用的推送机制之外的推送机制, 从第一次未命中目标 数据的用户的数据需求开始逐一分析,并根据相应的推送机制假设预推数据, 检验所述假设预推数据的命中率, 若所述命中率大于指定值, 则将该相应的 推送机制确定为与所述用户相关的推送机制。
为了解决上述问题, 本发明还提供了一种智能推送数据的装置, 包括: 获取模块, 设置为获取与用户相关的推送机制;
确定模块, 设置为根据所述推送机制确定出预推数据;
推送模块, 设置为将所述预推数据推送到指定的緩存。
优选地, 上述装置还具有下面特点: 所述推送机制包括:
推送用户指定的字段集内的数据的字段集机制、 推送用户访问过的表内 的数据的同表机制、 推送用户指定表集合内表的数据的表集合机制、 推送最 近使用过并被更新过的数据的最新机制或自动确定优先推送的数据的自动机 制。
优选地, 上述装置还具有下面特点: 所述获取模块, 具体设置为若判断 所述用户是第一次登录, 则获取推送用户指定的字段集内的数据的字段集机 制。
优选地, 上述装置还具有下面特点: 所述确定模块, 具体设置为根据所 述推送机制和用户的历史数据需求确定出所述预推数据。
优选地, 上述装置还具有下面特点: 还包括, 更新模块, 设置为接收到 用户的数据需求, 则更新用户的历史数据需求。
优选地, 上述装置还具有下面特点: 所述更新模块, 具体设置为若检测 所述数据需求不属于指定字段集内的字段, 则将用户该次访问的字段增加到 所述指定字段集; 或若检测所述数据需求不属于指定表集合内的字段, 则将 用户该次访问的表增加到所述指定表集合。
优选地, 上述装置还具有下面特点: 还包括,
分析模块, 设置为若判断根据用户的数据需求在所述緩存中连续未命中 目标数据的次数大于指定次数, 则对用户的历史数据需求进行分析, 重新确 定推送机制。
优选地, 上述装置还具有下面特点:
所述分析模块, 具体设置为若判断根据用户的数据需求在所述緩存中连 续未命中目标数据的次数大于指定次数, 则依次釆用除当前使用的推送机制 之外的推送机制,从第一次未命中目标数据的用户的数据需求开始逐一分析, 并根据相应的推送机制假设预推数据, 检验所述假设预推数据的命中率, 若 所述命中率大于指定值, 则将该相应的推送机制确定为与所述用户相关的推 送机制。
优选地, 上述装置还具有下面特点: 还包括, 配置模块, 设置为接受用 户配置的推送机制。
综上所述, 上述技术方案提供一种智能推送数据的方法和装置, 能够主 动预测并推送给客户端所需数据, 可以大大降低客户端未命中的数据记录, 大大提高客户端的平均响应时间, 并可使数据库和网络的负载更加均衡。 附图概述
图 1为本发明实施例的一种智能推送数据的装置;
图 2为本发明实施方式所提供的智能推送数据的方法的流程图。 本发明的较佳实施方式
下文中将结合附图对本发明的实施例进行详细说明。 需要说明的是, 在 不冲突的情况下, 本申请中的实施例及实施例中的特征可以相互任意组合。
图 1为本发明实施例的一种智能推送数据的装置, 如图 1所示, 本装置 主要包括: 选取模块、 确定模块和推送模块, 其中,
获取模块, 设置为获取与用户相关的推送机制;
确定模块, 设置为根据所述推送机制确定出预推数据;
推送模块, 设置为将所述预推数据推送到指定的緩存。
所述指定的緩存优选为所述用户的客户端中的緩存。
这样, 当客户端进行大量的数据查询时, 能够明显降低响应时间。
其中, 所述推送机制可以包括: 推送用户指定的字段集内的数据的字段 集机制、 推送用户访问过的表内的数据的同表机制、 推送用户指定表集合内 表的数据的表集合机制、 推送最近使用过并被更新过的数据的最新机制、 自 动确定优先推送的数据的自动机制等。
本实施例的装置还包括配置模块, 设置为为相应用户配置推送机制, 配 置时可以按用户组来配置, 即同一个用户组的用户配置同一个推送机制。 其中, 所述获取模块, 具体设置为若判断所述用户是第一次登录, 则获 取推送用户指定的字段集内的数据的字段集机制, 若用户的字段集合为空, 则不作处理, 等待用户的命令。
其中, 所述确定模块, 具体设置为根据所述推送机制和用户的历史数据 需求确定出所述预推数据。
本实施例的装置还可以包括更新模块, 设置为接收到用户的数据需求, 则更新用户的历史数据需求。
具体地, 更新模块若检测所述数据需求不属于指定字段集内的字段, 则 将用户该次访问的字段增加到所述指定字段集; 或若检测所述数据需求不属 于指定表集合内的字段, 则将用户该次访问的表增加到所述指定表集合。
本实施例的装置还可以包括分析模块, 设置为若判断根据用户的数据需 求在所述緩存中连续未命中目标数据的次数大于指定次数 (例如 10次, 指定 次数为可配置) , 则对用户的历史数据需求进行分析, 重新确定推送机制。
具体地, 分析模块若判断根据用户的数据需求在所述緩存中连续未命中 目标数据的次数大于指定次数, 则依次釆用除当前使用的推送机制之外的推 送机制, 从第一次未命中目标数据的用户的数据需求开始逐一分析, 并根据 相应的推送机制假设预推数据, 检验所述假设预推数据的命中率, 若所述命 中率大于指定值(例如 60% ) , 则将该相应的推送机制确定为与所述用户相 关的推送机制。 若命中率均未达到指定值, 则取消该用户的数据预推, 但优 选地, 分析模块可以继续预分析该用户的数据需求, 分析命中率, 当命中率 达到指定值时, 釆用相应的推送机制恢复对该用户的数据预推。
图 2为本发明实施方式所提供的智能推送数据的方法的流程图, 如图 2 所示, 本方法包括下面步骤:
S10、 获取与用户相关的推送机制;
S20、 根据所述推送机制确定出预推数据;
S30、 将所述预推数据推送到指定的缓存。 这样, 当客户端进行大量的数据查询时, 能够明显降低响应时间。
其中, 推送机制为分布式数据库釆用什么样的策略来选择主动推送给用 户的数据。 推送机制可以包括字段集机制、 同表机制、 表集合机制、 最新机 制、 自动机制等。
其中, 字段集机制表示优先推送用户指定的字段集内的数据;
同表机制表示优先推送用户访问过的表内的其他数据;
表集合机制表示优先推送用户指定表集合内表的数据;
最新机制表示优先考虑最近使用过并被更新过的数据;
自动机制表示自动确定(多机制混合)优先推送的数据。
其中, 推送机制配置可以按用户组来配置, 同一个用户组的用户可以配 置釆用同一种推送机制, 同一个用户属于不同的用户组时可以默认釆用所属 的第一个用户组配置的推送机制, 也可以在用户加入多用户组时指定推送机 制。
当用户登录时, 分布式数据库侧可以根据配置信息获取用户相关的推送 机制, 初始化系统并根据用户的历史数据需求确定出预推数据, 并推送到用 户客户端。
其中,用户的历史数据需求是用户上次登录分布式数据库时的数据需求, 则由分布式系统记录并序列化该数据需求, 在用户登录后重新启用, 大大降 氐了响应时间。
当用户第一次登录系统时, 釆用字段集机制推送字段集内的数据, 如果 用户的字段集合为空, 则不做处理, 等待用户指令。
在用户和系统交互时, 用户传递数据需求, 分布式数据库的客户端首先 查看緩存, 如果命中目标数据, 则直接返回结果, 并同时发送用户数据需求 到分布式数据库侧, 分布式数据库侧则更新用户的历史数据需求, 并根据推 送机制确定出用户预推数据, 并主动推送数据到用户的客户端。
如果客户端根据用户的数据需求在緩存中未命中目标数据, 则发送数据 需求到分布式数据库侧, 分布式数据库侧返回所需数据集, 并更新用户的历 史数据需求记录, 同时, 触发未命中数据需求事件, 并分析该次数据需求和 历史数据需求的关系。
其中,分布式数据库侧如果检测该次数据需求不属于指定字段集内字段, 则可以将用户该次访问的字段增加到所述指定字段集, 其中, 字段集中的字 段数目可配置。
分布式数据库侧如果检测该次数据需求不属于指定表集合内的字段, 则 可以将用户该次访问的表增加到所述指定表集合, 其中, 表集合中表的数目 可配置。
当分布式数据库侧判断根据用户的数据需求在所述緩存中连续未命中目 标数据的次数大于指定次数(如 10次, 可配置), 则需要对历史数据需求进 行分析, 并重新确定推送机制, 过程具体如下:
除去当前釆用的推送机制, 依次釆用其他各推送机制, 从第一次未命中 目标数据的用户的数据需求开始逐一分析, 并根据相应的推送机制假设预推 数据,并检验所述假设预推数据的命中率,当命中率大于指定值时(默认 60% , 可配置) , 则釆用此次的推送机制。 当所有推送机制均失效时, 即命中率均 未达到指定值, 则取消该用户的数据预推。 但同时, 分布式数据库侧依然会 预分析用户的数据需求, 当命中率达到指定值时, 则釆用相应的推送机制恢 复该用户的数据预推。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序 来指令相关硬件完成, 所述程序可以存储于计算机可读存储介质中, 如只读 存储器、 磁盘或光盘等。 可选地, 上述实施例的全部或部分步骤也可以使用 一个或多个集成电路来实现。 相应地, 上述实施例中的各模块 /单元可以釆用 硬件的形式实现, 也可以釆用软件功能模块的形式实现。 本发明的实施方式 不限制于任何特定形式的硬件和软件的结合。
以上仅为本发明的优选实施例, 当然, 本发明还可有其他多种实施例, 在不背离本发明精神及其实质的情况下, 熟悉本领域的技术人员当可根据本 发明的实施方式作出各种相应的改变和变形, 但这些相应的改变和变形都应 属于本申请所附的权利要求的保护范围。
工业实用性
上述技术方案提供一种智能推送数据的方法和装置, 以主动向客户端推 送数据, 可以大大降低客户端未命中的数据记录, 大大提高客户端的平均响 应时间, 提高客户端的响应速度, 并可使数据库和网络的负载更加均衡。

Claims

权 利 要 求 书
1、 一种智能推送数据的方法, 包括:
获取与用户相关的推送机制;
根据所述推送机制确定出预推数据;
将所述预推数据推送到指定的緩存。
2、 如权利要求 1所述的方法, 其中: 所述推送机制包括:
推送用户指定的字段集内的数据的字段集机制、 推送用户访问过的表内 的数据的同表机制、 推送用户指定表集合内表的数据的表集合机制、 推送最 近使用过并被更新过的数据的最新机制或自动确定优先推送的数据的自动机 制。
3、 如权利要求 1所述的方法, 其中:
所述获取与用户相关的推送机制的步骤包括: 若判断所述用户是第一次 登录, 则选取推送用户指定的字段集内的数据的字段集机制; 或者
所述根据所述推送机制确定出预推数据的步骤包括: 根据所述推送机制 和所述用户的历史数据需求确定出所述预推数据。
4、 如权利要求 3所述的方法, 其中: 在所述根据所述推送机制确定出预 推数据的步骤之前, 所述方法还包括:
若接收到所述用户的数据需求, 则更新所述用户的历史数据需求。
5、 如权利要求 4所述的方法, 其中: 在所述接收到所述用户的数据需求 的步骤之后, 所述方法还包括:
若检测所述数据需求不属于指定字段集内的字段, 则将所述用户该次访 问的字段增加到所述指定字段集; 或
若检测所述数据需求不属于指定表集合内的字段, 则将所述用户该次访 问的表增加到所述指定表集合。
6、 如权利要求 1所述的方法, 其中: 在所述将所述预推数据推送到指定 的緩存的步骤之后, 所述方法还包括:
若判断根据所述用户的数据需求在所述緩存中连续未命中目标数据的次 数大于指定次数, 则对所述用户的历史数据需求进行分析, 重新确定推送机 制。
7、 如权利要求 6所述的方法, 其中: 所述对所述用户的历史数据需求进 行分析, 重新确定推送机制的步骤包括:
依次釆用除当前使用的推送机制之外的推送机制, 从第一次未命中目标 数据的用户的数据需求开始逐一分析,并根据相应的推送机制假设预推数据, 检验所述假设预推数据的命中率, 若所述命中率大于指定值, 则将该相应的 推送机制确定为与所述用户相关的推送机制。
8、 一种智能推送数据的装置, 包括获取模块、 确定模块和推送模块, 其 中:
所述获取模块设置为: 获取与用户相关的推送机制;
所述确定模块设置为: 根据所述推送机制确定出预推数据;
所述推送模块设置为: 将所述预推数据推送到指定的緩存。
9、 如权利要求 8所述的装置, 其中: 所述推送机制包括:
推送用户指定的字段集内的数据的字段集机制、 推送用户访问过的表内 的数据的同表机制、 推送用户指定表集合内表的数据的表集合机制、 推送最 近使用过并被更新过的数据的最新机制或自动确定优先推送的数据的自动机 制。
10、 如权利要求 8所述的装置, 其中:
所述获取模块是设置为: 若判断所述用户是第一次登录, 则获取推送用 户指定的字段集内的数据的字段集机制; 或者
所述确定模块是设置为: 根据所述推送机制和用户的历史数据需求确定 出所述预推数据。
11、 如权利要求 8所述的装置, 其还包括:
更新模块, 其设置为: 若接收到所述用户的数据需求, 则更新该用户的 历史数据需求。
12、 如权利要求 11所述的装置, 其中: 所述更新模块是设置为: 若检测所述数据需求不属于指定字段集内的字 段, 则将所述用户该次访问的字段增加到所述指定字段集; 或若检测所述数 据需求不属于指定表集合内的字段, 则将所述用户该次访问的表增加到所述 指定表集合。
13、 如权利要求 8所述的装置, 其还包括,
分析模块, 其设置为: 若判断根据所述用户的数据需求在所述緩存中连 续未命中目标数据的次数大于指定次数, 则对该用户的历史数据需求进行分 析, 重新确定推送机制。
14、 如权利要求 13所述的装置, 其中:
所述分析模块是设置为: 若判断根据所述用户的数据需求在所述緩存中 连续未命中目标数据的次数大于指定次数, 则依次釆用除当前使用的推送机 制之外的推送机制, 从第一次未命中目标数据的用户的数据需求开始逐一分 析, 并根据相应的推送机制假设预推数据,检验所述假设预推数据的命中率, 若所述命中率大于指定值, 则将该相应的推送机制确定为与所述用户相关的 推送机制。
15、 如权利要求 8-14任一项所述的装置, 其还包括,
配置模块, 其设置为: 为相应的用户配置推送机制。
PCT/CN2011/080098 2010-11-18 2011-09-23 一种智能推送数据的方法和装置 WO2012065478A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2010105492254A CN102467553A (zh) 2010-11-18 2010-11-18 一种智能推送数据的方法和装置
CN201010549225.4 2010-11-18

Publications (1)

Publication Number Publication Date
WO2012065478A1 true WO2012065478A1 (zh) 2012-05-24

Family

ID=46071196

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/080098 WO2012065478A1 (zh) 2010-11-18 2011-09-23 一种智能推送数据的方法和装置

Country Status (2)

Country Link
CN (1) CN102467553A (zh)
WO (1) WO2012065478A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103020245B (zh) * 2012-12-19 2015-10-07 北京奇虎科技有限公司 用于下载数据的方法和装置
CN103249006B (zh) * 2013-04-28 2015-09-30 清华大学 一种基于多播的网络数据预推送方法
CN104333594B (zh) * 2014-11-05 2018-07-27 无锡成电科大科技发展有限公司 基于光传输网络的云平台资源采集加速方法和系统
CN105868207B (zh) * 2015-01-21 2019-10-15 方正宽带网络服务有限公司 一种网络资源推送方法及装置
CN108093076A (zh) * 2017-12-29 2018-05-29 北京奇虎科技有限公司 消息推送方法和装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1653453A (zh) * 2002-03-11 2005-08-10 捷讯研究有限公司 用于将数据推到移动设备的系统和方法
US20050240652A1 (en) * 2004-04-21 2005-10-27 International Business Machines Corporation Application Cache Pre-Loading
CN101105795A (zh) * 2006-10-27 2008-01-16 北京搜神网络技术有限责任公司 基于网络行为的个性化推荐方法和系统
US20090193099A1 (en) * 2008-01-29 2009-07-30 Palo Alto Research Center Incorporated Method and apparatus for automatically incorporating hypothetical context information into recommendation queries

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101042709A (zh) * 2007-04-11 2007-09-26 芦树鹏 主动式搜索
CN101661474B (zh) * 2008-08-26 2012-07-04 华为技术有限公司 一种搜索方法和系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1653453A (zh) * 2002-03-11 2005-08-10 捷讯研究有限公司 用于将数据推到移动设备的系统和方法
US20050240652A1 (en) * 2004-04-21 2005-10-27 International Business Machines Corporation Application Cache Pre-Loading
CN101105795A (zh) * 2006-10-27 2008-01-16 北京搜神网络技术有限责任公司 基于网络行为的个性化推荐方法和系统
US20090193099A1 (en) * 2008-01-29 2009-07-30 Palo Alto Research Center Incorporated Method and apparatus for automatically incorporating hypothetical context information into recommendation queries

Also Published As

Publication number Publication date
CN102467553A (zh) 2012-05-23

Similar Documents

Publication Publication Date Title
US11860874B2 (en) Multi-partitioning data for combination operations
US11494380B2 (en) Management of distributed computing framework components in a data fabric service system
US11645210B2 (en) Cache aware searching based on one or more files in remote storage
US11636105B2 (en) Generating a subquery for an external data system using a configuration file
US9626217B2 (en) Information processing apparatus, information processing method, recording medium and information processing system
US11921672B2 (en) Query execution at a remote heterogeneous data store of a data fabric service
US11151137B2 (en) Multi-partition operation in combination operations
CN109831500B (zh) Kubernetes集群中配置文件与Pod的同步方法
US20190147085A1 (en) Converting and modifying a subquery for an external data system
CN100507855C (zh) 资源优化组件和方法
US9665603B2 (en) Row-based data filtering at a database level
WO2012065478A1 (zh) 一种智能推送数据的方法和装置
WO2020027867A1 (en) Generating a subquery for a distinct data intake and query system
CN103685590B (zh) 获取ip地址的方法及系统
EP3058481B1 (en) Acceleration based on cached flows
CN106254561B (zh) 一种网络资源文件的实时离线下载方法及系统
JP5950267B2 (ja) データベース管理装置、データベース管理方法及び記憶媒体
CN113474764A (zh) 共享数据库对象上的流
CN109937566A (zh) 用于在联网环境中进行计算卸载的方法和装置
US7730164B1 (en) Bootstrap approaches to downloading data in response to a download indication
CN107301084B (zh) 集群服务器的虚拟机迁移方法、装置、服务器及存储介质
US11604789B1 (en) Bi-directional query updates in a user interface
US8990169B2 (en) Statistics collection for database tables
US20130212101A1 (en) Portlet processing apparatus, portal server, portal system, portlet processing method and recording medium
CN109684611A (zh) 一种字典转码方法、装置、存储介质及终端

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11841170

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11841170

Country of ref document: EP

Kind code of ref document: A1