技术领域technical field
本发明涉及负载均衡调度技术领域，具体是一种集群自反馈式负载均衡调度系统及方法。The invention relates to the technical field of load balancing scheduling, in particular to a cluster self-feedback load balancing scheduling system and method.
背景技术Background technique
随着互联网技术的发展，单台服务器组成的系统由于中央处理器、输入输出处理速度等方面的限制，已经不能满足快速增长的数据量及业务处理的需求。集群技术的出现，可以在付出较低成本的情况下获得在性能、可靠性、灵活性方面的相对较高的收益，但是关于集群调度这一核心问题现有的机制还不够完善，存在或多或少的问题。比如：没有考虑是否已有集群处理过类似请求问题，没有比较各集群处理相同请求的效率问题和没有实现调度服务器和集群之间的动态协调沟通的问题等。With the development of Internet technology, a system composed of a single server can no longer meet the needs of rapidly growing data volume and business processing due to the limitations of the central processing unit and input and output processing speed. The emergence of cluster technology can obtain relatively high benefits in terms of performance, reliability, and flexibility at a lower cost. However, the existing mechanism for the core issue of cluster scheduling is not perfect enough, and there are more or less or less question. For example, it does not consider whether there are clusters that have processed similar requests, does not compare the efficiency of each cluster in processing the same request, and does not realize the dynamic coordination and communication between the scheduling server and the cluster, etc.
因此，集群间负载均衡的实现没有完全高效利用可用资源，存在一定的分配不均的现象。Therefore, the implementation of inter-cluster load balancing does not fully and efficiently utilize available resources, and there is a certain phenomenon of uneven distribution.
发明内容Contents of the invention
为了克服上述现有技术的缺点，本发明提供了一种集群自反馈式负载均衡调度系统及方法，真正实现任务分配的负载均衡。In order to overcome the above-mentioned shortcomings of the prior art, the present invention provides a cluster self-feedback load balancing scheduling system and method to truly realize load balancing of task assignment.
一种集群自反馈式负载均衡调度方法，调度中心服务器按表查询机制实现各负载均衡以及反馈式更新各表状态；A cluster self-feedback load balancing scheduling method, in which the scheduling center server implements load balancing according to the table query mechanism and updates the status of each table in a feedback manner;
(1)当调度中心服务器接收到用户请求后，首先查看资源分布表，根据资源分布表中内容，确定可处理此请求的集群的范围；(1) When the dispatch center server receives the user request, it first checks the resource distribution table, and determines the range of clusters that can handle the request according to the content in the resource distribution table;
(2)确定集群范围后，接着查询空闲表，根据选定的集群范围依次查看每个可响应集群所剩余的资源情况；(2) After determining the cluster range, then query the free list, and check the remaining resources of each responding cluster in turn according to the selected cluster range;
若存在多个所剩资源较多的集群则接着查询分配历史记录表,看是否存在已经执行过同类请求消息的记录；若存在，则比较符合条件的集群上次处理此类似消息所占用的资源情况和最终完成所消耗的时间，两者结合优胜者获得处理该请求的权利；若不存在，则从多个符合条件的集群中选出所剩资源最多的获得处理该请求的权利；If there are multiple clusters with more remaining resources, then query the allocation history table to see if there is a record that has executed the same type of request message; if it exists, compare the resources occupied by the qualified cluster for processing this similar message last time The situation and the time consumed for final completion, the winner who combines the two will get the right to process the request; if it does not exist, select the one with the most remaining resources from multiple eligible clusters to get the right to process the request;
若可响应集群皆无所剩资源，则将请求放入排队等候表中，直至空闲表中出现可分配资源后，跳出排队等候表接着查询空闲表；If there are no remaining resources in the responding cluster, the request will be placed in the queue waiting list until the available resources appear in the free list, jump out of the queue waiting list and then query the free list;
(3)调度中心服务器给最终选定的集群分配用户请求任务，集群开始处理该请求；(3) The dispatch center server assigns user request tasks to the finally selected cluster, and the cluster starts to process the request;
(4)集群及时反馈集群状态更新调度中心服务器的状态，各表项相互结合关联，实现反馈式的负载均衡任务调度分配。(4) The cluster feeds back the status of the cluster in a timely manner to update the status of the scheduling center server, and the table items are combined and correlated with each other to realize the feedback load balancing task scheduling assignment.
优选的，各集群中的Master服务器利用以下两种报文实时通知调度中心服务器更新表中各集群状态：周期发送Hello报文，通告给调度中心服务器自身状态；并且当新建或完成处理任务时，实时发送Update报文通知调度中心服务器修改各表状态；Master服务器是每个集群中存在的一台管理本集群其它服务器的设备。Preferably, the Master server in each cluster utilizes the following two kinds of messages to notify the dispatching center server of each cluster state in the update table in real time: periodically send the Hello message to notify the dispatching center server of its own state; and when creating a new or completing a processing task, Send Update messages in real time to notify the dispatching center server to modify the status of each table; the Master server is a device in each cluster that manages other servers in the cluster.
优选的，两种报文的工作方式如下：Preferably, the working modes of the two messages are as follows:
(1)在集群稳定的处理用户请求消息的过程中，集群中的Master服务器会以60秒为一个周期，按时发送Hello报文，通告本集群各资源使用情况给调度中心服务器；(1) When the cluster is stably processing user request messages, the Master server in the cluster will send a Hello message on time with 60 seconds as a cycle, and notify the resource usage of the cluster to the dispatching center server;
(2)调度中心服务器收到Hello报文后将比对各表中的数值，发现没有变化，将忽略此报文；(2) After receiving the Hello message, the dispatching center server will compare the values in each table and find that there is no change, and will ignore this message;
(3)当集群收到一个新的用户请求消息或者处理完成一个请求消息后，将立即发送一个Update报文，里面放置释放或新占用的资源情况；(3) When the cluster receives a new user request message or completes processing a request message, it will immediately send an Update message, which contains the released or newly occupied resources;
(4)调度中心服务器收到此Update报文后，依次比对负载表、空闲表中的信息并做出相应的修改；同时，在分配历史记录表中添加已完成的请求消息的记录和完成后所涉及的一系列参数；(4) After the dispatching center server receives the Update message, it compares the information in the load list and the free list in turn and makes corresponding modifications; at the same time, it adds the record and completion of the completed request message in the distribution history record table. A series of parameters involved in the latter;
(5)调度中心服务器的各表项值得到及时更新。(5) The value of each table item of the dispatching center server is updated in time.
优选的，调度中心服务器中具有五张表，调度中心服务器按表查询机制如下：Preferably, there are five tables in the dispatch center server, and the dispatch center server queries the table as follows:
资源分布表：存放各集群可以处理请求信息的集合；Resource distribution table: stores the collection of request information that each cluster can process;
负载表：存放各集群正在处理的请求信息的集合；Load table: stores the collection of request information being processed by each cluster;
空闲表：存放各集群所剩资源情况的集合；Free list: store the collection of remaining resources of each cluster;
分配历史记录表：存放各集群已处理过请求消息的种类及处理各请求所占用资源情况和最终完成所消耗的时间；Allocation history table: store the types of request messages processed by each cluster, the resources occupied by processing each request, and the time consumed for final completion;
排队等候表：存放等待处理的请求信息。Queue waiting table: store request information waiting to be processed.
一种集群自反馈式负载均衡调度系统，调度中心服务器按表查询机制实现各负载均衡以及反馈式更新各表状态；A cluster self-feedback load balancing scheduling system, in which the scheduling center server implements load balancing and updates the status of each table according to the table query mechanism;
查询模块；调度中心服务器接收到用户请求后，首先查看资源分布表，根据资源分布表中内容，确定可处理此请求的集群的范围；Query module; after receiving the user request, the scheduling center server first checks the resource distribution table, and determines the range of clusters that can handle the request according to the content in the resource distribution table;
选择模块：确定集群范围后，接着查询空闲表，根据选定的集群范围依次查看每个可响应集群所剩余的资源情况；Selection module: After determining the range of the cluster, then query the free list, and check the remaining resources of each responding cluster in turn according to the selected cluster range;
若存在多个所剩资源较多的集群则接着查询分配历史记录表，看是否存在已经执行过同类请求消息的记录；若存在，则比较符合条件的集群上次处理此类似消息所占用的资源情况和最终完成所消耗的时间，两者结合优胜者获得处理该请求的权利；若不存在，则从多个符合条件的集群中选出所剩资源最多的获得处理该请求的权利；If there are multiple clusters with more remaining resources, then query the allocation history table to see if there is a record that has executed the same type of request message; if there is, compare the resources occupied by the qualified cluster for processing this similar message last time The situation and the time consumed for final completion, the winner who combines the two will get the right to process the request; if it does not exist, select the one with the most remaining resources from multiple eligible clusters to get the right to process the request;
若可响应集群皆无所剩资源，则将请求放入排队等候表中，直至空闲表中出现可分配资源后，跳出排队等候表接着查询空闲表；If there are no remaining resources in the responding cluster, the request will be placed in the queue waiting list until the available resources appear in the free list, jump out of the queue waiting list and then query the free list;
执行模块：调度中心服务器给最终选定的集群分配用户请求任务，集群开始处理该请求；Execution module: the scheduling center server assigns user request tasks to the finally selected cluster, and the cluster starts to process the request;
反馈模块：集群及时反馈集群状态更新调度中心服务器的状态，各表项相互结合关联，实现反馈式的负载均衡任务调度分配。Feedback module: The cluster feeds back the status of the cluster in a timely manner to update the status of the scheduling center server, and the table items are associated with each other to realize the feedback load balancing task scheduling and distribution.
本发明的有益效果是：通过调度中心服务器中各表之间的查询关联，引入历史机制查询，在处理类似请求时更加高效；处理新增请求时保证切实高效的负载均衡的实现。同时，反馈机制的应用，为切实负载均衡的实现增加了进一步保障。The beneficial effect of the present invention is: through the query correlation between tables in the dispatching center server, the history mechanism query is introduced, which is more efficient when processing similar requests; and the realization of effective and efficient load balancing is ensured when processing new requests. At the same time, the application of the feedback mechanism adds a further guarantee for the realization of practical load balancing.
附图说明Description of drawings
下面结合附图对本发明作进一步说明。The present invention will be further described below in conjunction with accompanying drawing.
图1为发明中的整体架构示意图；Fig. 1 is a schematic diagram of the overall architecture in the invention;
图2为调度中心服务器中表的分布；Fig. 2 is the distribution of table in the dispatching center server;
图3为负载均衡工作流程示意图；Figure 3 is a schematic diagram of the load balancing workflow;
图4为调节系统原理框图。Figure 4 is a block diagram of the regulation system.
具体实施方式detailed description
如图1所示，在所有集群的“上方”设置了一台调度中心服务器，用来接收用户发来的请求信息，并且通过其各项机制的设定来实现负载均衡。调度中心服务器就像是用户与集群之间的一个桥梁，起到了接收、调度、分配等一系列功能，使用户和集群之间忽略其他不必要因素的影响，看起来就像只与调度中心服务器进行通信一样。As shown in Figure 1, a dispatch center server is set up "on top" of all clusters to receive request information from users and achieve load balancing through the settings of its various mechanisms. The dispatch center server is like a bridge between the user and the cluster. It performs a series of functions such as receiving, dispatching, and distribution, so that the influence of other unnecessary factors is ignored between the user and the cluster. It looks like it is only connected with the dispatch center server. Same as communicating.
如图2所示，调度中心服务器五张表之间相互协商工作，保证负载均衡的高效运行，各个存放内容如下：As shown in Figure 2, the five tables of the dispatching center server negotiate with each other to ensure the efficient operation of load balancing. The contents of each storage are as follows:
资源分布表：存放各集群可以处理请求信息的集合；Resource distribution table: stores the collection of request information that each cluster can process;
负载表：存放各集群正在处理的请求信息的集合；Load table: stores the collection of request information being processed by each cluster;
空闲表：存放各集群所剩资源情况的集合；Free list: store the collection of remaining resources of each cluster;
分配历史记录表：存放各集群已处理过请求消息的种类及处理各请求所占用资源情况和最终完成所消耗的时间；Allocation history table: store the types of request messages processed by each cluster, the resources occupied by processing each request, and the time consumed for final completion;
排队等候表：存放等待处理的请求信息。Queue waiting table: store request information waiting to be processed.
如图3所示，一种集群自反馈式负载均衡调度方法，调度中心服务器按表查询机制实现各负载均衡以及反馈式更新各表状态；As shown in Figure 3, a cluster self-feedback load balancing scheduling method, the scheduling center server realizes each load balancing and feedback updates the state of each table according to the table query mechanism;
(1)当调度中心服务器接收到用户请求后，首先查看资源分布表，根据资源分布表中内容，确定可处理此请求的集群的范围；(1) When the dispatch center server receives the user request, it first checks the resource distribution table, and determines the range of clusters that can handle the request according to the content in the resource distribution table;
(2)确定集群范围后，接着查询空闲表，根据选定的集群范围依次查看每个可响应集群所剩余的资源情况；(2) After determining the cluster range, then query the free list, and check the remaining resources of each responding cluster in turn according to the selected cluster range;
若存在多个所剩资源较多的集群则接着查询分配历史记录表,看是否存在已经执行过同类请求消息的记录；若存在，则比较符合条件的集群上次处理此类似消息所占用的资源情况和最终完成所消耗的时间，两者结合优胜者获得处理该请求的权利；若不存在，则从多个符合条件的集群中选出所剩资源最多的获得处理该请求的权利；If there are multiple clusters with more remaining resources, then query the allocation history table to see if there is a record that has executed the same type of request message; if it exists, compare the resources occupied by the qualified cluster for processing this similar message last time The situation and the time consumed for final completion, the winner who combines the two will get the right to process the request; if it does not exist, select the one with the most remaining resources from multiple eligible clusters to get the right to process the request;
若可响应集群皆无所剩资源，则将请求放入排队等候表中，直至空闲表中出现可分配资源后，跳出排队等候表接着查询空闲表；If there are no remaining resources in the responding cluster, the request will be placed in the queue waiting list until the available resources appear in the free list, jump out of the queue waiting list and then query the free list;
(3)调度中心服务器给最终选定的集群分配用户请求任务，集群开始处理该请求；(3) The dispatch center server assigns user request tasks to the finally selected cluster, and the cluster starts to process the request;
(4)集群及时反馈集群状态更新调度中心服务器的状态，各表项相互结合关联，实现反馈式的负载均衡任务调度分配。(4) The cluster feeds back the status of the cluster in a timely manner to update the status of the scheduling center server, and the table items are combined and correlated with each other to realize the feedback load balancing task scheduling assignment.
各集群中的Master服务器利用以下两种报文实时通知调度中心服务器更新表中各集群状态，Master服务器是每个集群中存在的一台管理本集群其它服务器的设备。The Master server in each cluster uses the following two messages to notify the dispatching center server to update the status of each cluster in the table in real time. The Master server is a device in each cluster that manages other servers in the cluster.
Hello报文：周期性发送，里面包含的内容有：集群的CPU使用率、内存占用情况等一系列资源；Hello message: sent periodically, which contains a series of resources such as the cluster's CPU usage and memory usage;
Update报文：触发式更新，当集群新接收或刚处理结束请求时，随即向调度中心服务器进行通告；里面包含的内容有：新占用或解除占用的CPU使用量、占用或释放内存的大小等一系列资源。Update message: Triggered update, when the cluster receives or just finishes processing a request, it immediately notifies the dispatching center server; the content contained in it includes: newly occupied or released CPU usage, the size of occupied or released memory, etc. A range of resources.
两种报文的工作方式如下：The two telegrams work as follows:
(1)在集群稳定的处理用户请求消息的过程中，集群中的Master服务器会以60秒为一个周期，按时发送Hello报文，通告本集群各资源使用情况给调度中心服务器；(1) When the cluster is stably processing user request messages, the Master server in the cluster will send a Hello message on time with 60 seconds as a cycle, and notify the resource usage of the cluster to the dispatching center server;
(2)调度中心服务器收到Hello报文后将比对各表中的数值，发现没有变化，将忽略此报文；(2) After receiving the Hello message, the dispatching center server will compare the values in each table and find that there is no change, and will ignore this message;
(3)当集群收到一个新的用户请求消息或者处理完成一个请求消息后，将立即发送一个Update报文，里面放置释放或新占用的资源情况；(3) When the cluster receives a new user request message or completes processing a request message, it will immediately send an Update message, which contains the released or newly occupied resources;
(4)调度中心服务器收到此Update报文后，依次比对负载表、空闲表中的信息并做出相应的修改；同时，在分配历史记录表中添加已完成的请求消息的记录和完成后所涉及的一系列参数；(4) After the dispatching center server receives the Update message, it compares the information in the load list and the free list in turn and makes corresponding modifications; at the same time, it adds the record and completion of the completed request message in the distribution history record table. A series of parameters involved in the latter;
(5)调度中心服务器的各表项值得到及时更新。(5) The value of each table item of the dispatching center server is updated in time.
如图4所示，一种集群自反馈式负载均衡调度系统，调度中心服务器按表查询机制实现各负载均衡以及反馈式更新各表状态；As shown in Figure 4, a cluster self-feedback load balancing scheduling system, the scheduling center server realizes each load balancing and feedback updates the state of each table according to the table query mechanism;
查询模块；调度中心服务器接收到用户请求后，首先查看资源分布表，根据资源分布表中内容，确定可处理此请求的集群的范围；Query module; after receiving the user request, the scheduling center server first checks the resource distribution table, and determines the range of clusters that can handle the request according to the content in the resource distribution table;
选择模块：确定集群范围后，接着查询空闲表，根据选定的集群范围依次查看每个可响应集群所剩余的资源情况；Selection module: After determining the range of the cluster, then query the free list, and check the remaining resources of each responding cluster in turn according to the selected cluster range;
若存在多个所剩资源较多的集群则接着查询分配历史记录表,看是否存在已经执行过同类请求消息的记录；若存在，则比较符合条件的集群上次处理此类似消息所占用的资源情况和最终完成所消耗的时间，两者结合优胜者获得处理该请求的权利；若不存在，则从多个符合条件的集群中选出所剩资源最多的获得处理该请求的权利；If there are multiple clusters with more remaining resources, then query the allocation history table to see if there is a record that has executed the same type of request message; if it exists, compare the resources occupied by the qualified cluster for processing this similar message last time The situation and the time consumed for final completion, the winner who combines the two will get the right to process the request; if it does not exist, select the one with the most remaining resources from multiple eligible clusters to get the right to process the request;
若可响应集群皆无所剩资源，则将请求放入排队等候表中，直至空闲表中出现可分配资源后，跳出排队等候表接着查询空闲表；If there are no remaining resources in the responding cluster, the request will be placed in the queue waiting list until the available resources appear in the free list, jump out of the queue waiting list and then query the free list;
执行模块：调度中心服务器给最终选定的集群分配用户请求任务，集群开始处理该请求；Execution module: the scheduling center server assigns user request tasks to the finally selected cluster, and the cluster starts to process the request;
反馈模块：集群及时反馈集群状态更新调度中心服务器的状态，各表项相互结合关联，实现反馈式的负载均衡任务调度分配。Feedback module: The cluster feeds back the status of the cluster in a timely manner to update the status of the scheduling center server, and the table items are associated with each other to realize the feedback load balancing task scheduling and distribution.
调度中心服务器接收到用户请求后，查看自身内放置的资源分布表，确定相关可处理集群，进一步结合空闲表和分配历史记录表做出最合理的处理方案；各集群中的Master服务器周期发送Hello报文，通告给调度中心服务器自身状态；并且当新建或完成处理任务时，实时发送Update报文通知调度中心服务器修改各表状态，从而实现反馈式的负载均衡任务调度分配。After receiving the user request, the scheduling center server checks the resource distribution table placed in itself, determines the relevant clusters that can be processed, and further combines the free list and allocation history table to make the most reasonable processing plan; the Master server in each cluster periodically sends Hello message to inform the scheduling center server of its own status; and when a new or completed processing task is created, an Update message is sent in real time to notify the scheduling center server to modify the status of each table, thereby realizing feedback load balancing task scheduling and allocation.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610424982.6ACN105959395B (en) | 2016-06-15 | 2016-06-15 | A cluster self-feedback load balancing scheduling system and method |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610424982.6ACN105959395B (en) | 2016-06-15 | 2016-06-15 | A cluster self-feedback load balancing scheduling system and method |
| Publication Number | Publication Date |
|---|---|
| CN105959395Atrue CN105959395A (en) | 2016-09-21 |
| CN105959395B CN105959395B (en) | 2019-04-19 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610424982.6AExpired - Fee RelatedCN105959395B (en) | 2016-06-15 | 2016-06-15 | A cluster self-feedback load balancing scheduling system and method |
| Country | Link |
|---|---|
| CN (1) | CN105959395B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106886452A (en)* | 2017-01-23 | 2017-06-23 | 北京思特奇信息技术股份有限公司 | A kind of method of simplified cloud system task scheduling |
| CN107172193A (en)* | 2017-06-20 | 2017-09-15 | 郑州云海信息技术有限公司 | A kind of load-balancing method and its device based on cluster |
| CN107547616A (en)* | 2017-05-27 | 2018-01-05 | 新华三技术有限公司 | AP load-balancing methods and device |
| CN108874538A (en)* | 2018-05-31 | 2018-11-23 | 合肥本源量子计算科技有限责任公司 | It is a kind of for dispatching the dispatch server, dispatching method and application of quantum computer |
| CN109040283A (en)* | 2018-08-23 | 2018-12-18 | 上海海事大学 | A kind of modified load-balancing algorithm based on difference reaction type |
| CN109032803A (en)* | 2018-08-01 | 2018-12-18 | 阿里巴巴集团控股有限公司 | Data processing method and device, client |
| CN109145053A (en)* | 2018-08-01 | 2019-01-04 | 阿里巴巴集团控股有限公司 | Data processing method and device, client, server |
| CN109347966A (en)* | 2018-10-31 | 2019-02-15 | 许继集团有限公司 | A server cluster communication method, terminal device and communication server |
| CN109460293A (en)* | 2018-10-11 | 2019-03-12 | 东南大学 | Computing resource selection method in wireless cloud computing system under distributed computing environment |
| CN109902919A (en)* | 2019-01-17 | 2019-06-18 | 平安城市建设科技(深圳)有限公司 | Server assets management method, device, equipment and readable storage medium storing program for executing |
| CN111176697A (en)* | 2020-01-02 | 2020-05-19 | 广州虎牙科技有限公司 | Service instance deployment method, data processing method and cluster federation |
| CN112445575A (en)* | 2020-11-27 | 2021-03-05 | 中国工商银行股份有限公司 | Multi-cluster resource scheduling method, device and system |
| CN112584193A (en)* | 2020-12-24 | 2021-03-30 | 杭州米络星科技(集团)有限公司 | Method for constructing real-time streaming media cluster scheduling by using UDP (user Datagram protocol) characteristics |
| CN112751945A (en)* | 2021-04-02 | 2021-05-04 | 人民法院信息技术服务中心 | Method, device, equipment and storage medium for realizing distributed cloud service |
| CN113010309A (en)* | 2021-03-02 | 2021-06-22 | 北京达佳互联信息技术有限公司 | Cluster resource scheduling method, device, storage medium, equipment and program product |
| CN113366399A (en)* | 2019-12-30 | 2021-09-07 | 深圳元戎启行科技有限公司 | Vehicle control method and device based on remote takeover and computer equipment |
| CN113535851A (en)* | 2019-03-28 | 2021-10-22 | 北京忆芯科技有限公司 | Update and query of distributed KV storage system |
| CN113934525A (en)* | 2021-10-13 | 2022-01-14 | 正数网络技术有限公司 | Hadoop cluster task scheduling method based on positive and negative feedback load scheduling algorithm |
| CN118012608A (en)* | 2024-01-23 | 2024-05-10 | 朴道征信有限公司 | Data source call load balancing method, device, electronic equipment and medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7587437B2 (en)* | 2002-07-22 | 2009-09-08 | Fujitsu Limited | Parallel efficiency calculation method and apparatus |
| CN101957780A (en)* | 2010-08-17 | 2011-01-26 | 中国电子科技集团公司第二十八研究所 | Resource state information-based grid task scheduling processor and grid task scheduling processing method |
| CN103226467A (en)* | 2013-05-23 | 2013-07-31 | 中国人民解放军国防科学技术大学 | Data parallel processing method and system as well as load balancing scheduler |
| CN104572305A (en)* | 2015-01-26 | 2015-04-29 | 赞奇科技发展有限公司 | Load-balanced cluster rendering task scheduling method |
| CN104699736A (en)* | 2014-11-21 | 2015-06-10 | 北京华悦科技有限公司 | Distributed massive data acquisition system and method based on mobile devices |
| CN105589704A (en)* | 2014-10-22 | 2016-05-18 | 北京云巢动脉科技有限公司 | Method and system for accelerating virtual machine startup |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7587437B2 (en)* | 2002-07-22 | 2009-09-08 | Fujitsu Limited | Parallel efficiency calculation method and apparatus |
| CN101957780A (en)* | 2010-08-17 | 2011-01-26 | 中国电子科技集团公司第二十八研究所 | Resource state information-based grid task scheduling processor and grid task scheduling processing method |
| CN103226467A (en)* | 2013-05-23 | 2013-07-31 | 中国人民解放军国防科学技术大学 | Data parallel processing method and system as well as load balancing scheduler |
| CN105589704A (en)* | 2014-10-22 | 2016-05-18 | 北京云巢动脉科技有限公司 | Method and system for accelerating virtual machine startup |
| CN104699736A (en)* | 2014-11-21 | 2015-06-10 | 北京华悦科技有限公司 | Distributed massive data acquisition system and method based on mobile devices |
| CN104572305A (en)* | 2015-01-26 | 2015-04-29 | 赞奇科技发展有限公司 | Load-balanced cluster rendering task scheduling method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106886452A (en)* | 2017-01-23 | 2017-06-23 | 北京思特奇信息技术股份有限公司 | A kind of method of simplified cloud system task scheduling |
| CN106886452B (en)* | 2017-01-23 | 2020-08-18 | 北京思特奇信息技术股份有限公司 | Method for simplifying task scheduling of cloud system |
| CN107547616A (en)* | 2017-05-27 | 2018-01-05 | 新华三技术有限公司 | AP load-balancing methods and device |
| CN107172193A (en)* | 2017-06-20 | 2017-09-15 | 郑州云海信息技术有限公司 | A kind of load-balancing method and its device based on cluster |
| CN108874538A (en)* | 2018-05-31 | 2018-11-23 | 合肥本源量子计算科技有限责任公司 | It is a kind of for dispatching the dispatch server, dispatching method and application of quantum computer |
| CN108874538B (en)* | 2018-05-31 | 2020-10-13 | 合肥本源量子计算科技有限责任公司 | Scheduling server, scheduling method and application method for scheduling quantum computer |
| CN109145053A (en)* | 2018-08-01 | 2019-01-04 | 阿里巴巴集团控股有限公司 | Data processing method and device, client, server |
| US11233878B2 (en) | 2018-08-01 | 2022-01-25 | Advanced New Technologies Co., Ltd. | Data processing method, apparatus, and client device |
| US11563805B2 (en) | 2018-08-01 | 2023-01-24 | Advanced New Technologies Co., Ltd. | Method, apparatus, client terminal, and server for data processing |
| CN109032803A (en)* | 2018-08-01 | 2018-12-18 | 阿里巴巴集团控股有限公司 | Data processing method and device, client |
| CN109040283A (en)* | 2018-08-23 | 2018-12-18 | 上海海事大学 | A kind of modified load-balancing algorithm based on difference reaction type |
| CN109460293A (en)* | 2018-10-11 | 2019-03-12 | 东南大学 | Computing resource selection method in wireless cloud computing system under distributed computing environment |
| CN109460293B (en)* | 2018-10-11 | 2022-01-28 | 东南大学 | Computing resource selection method under distributed computing environment in wireless cloud computing system |
| CN109347966A (en)* | 2018-10-31 | 2019-02-15 | 许继集团有限公司 | A server cluster communication method, terminal device and communication server |
| CN109347966B (en)* | 2018-10-31 | 2021-08-03 | 许继集团有限公司 | A server cluster communication method, terminal device and communication server |
| CN109902919A (en)* | 2019-01-17 | 2019-06-18 | 平安城市建设科技(深圳)有限公司 | Server assets management method, device, equipment and readable storage medium storing program for executing |
| CN113535851A (en)* | 2019-03-28 | 2021-10-22 | 北京忆芯科技有限公司 | Update and query of distributed KV storage system |
| CN113366399B (en)* | 2019-12-30 | 2024-09-20 | 深圳元戎启行科技有限公司 | Vehicle control method and device based on remote take-over and computer equipment |
| CN113366399A (en)* | 2019-12-30 | 2021-09-07 | 深圳元戎启行科技有限公司 | Vehicle control method and device based on remote takeover and computer equipment |
| CN111176697B (en)* | 2020-01-02 | 2024-02-13 | 广州虎牙科技有限公司 | Service instance deployment method, data processing method and cluster federation |
| CN111176697A (en)* | 2020-01-02 | 2020-05-19 | 广州虎牙科技有限公司 | Service instance deployment method, data processing method and cluster federation |
| CN112445575B (en)* | 2020-11-27 | 2024-01-26 | 中国工商银行股份有限公司 | Multi-cluster resource scheduling method, device and system |
| CN112445575A (en)* | 2020-11-27 | 2021-03-05 | 中国工商银行股份有限公司 | Multi-cluster resource scheduling method, device and system |
| CN112584193A (en)* | 2020-12-24 | 2021-03-30 | 杭州米络星科技(集团)有限公司 | Method for constructing real-time streaming media cluster scheduling by using UDP (user Datagram protocol) characteristics |
| CN113010309B (en)* | 2021-03-02 | 2022-10-25 | 北京达佳互联信息技术有限公司 | Cluster resource scheduling method, device, storage medium, equipment and program product |
| CN113010309A (en)* | 2021-03-02 | 2021-06-22 | 北京达佳互联信息技术有限公司 | Cluster resource scheduling method, device, storage medium, equipment and program product |
| CN112751945A (en)* | 2021-04-02 | 2021-05-04 | 人民法院信息技术服务中心 | Method, device, equipment and storage medium for realizing distributed cloud service |
| CN113934525A (en)* | 2021-10-13 | 2022-01-14 | 正数网络技术有限公司 | Hadoop cluster task scheduling method based on positive and negative feedback load scheduling algorithm |
| CN118012608A (en)* | 2024-01-23 | 2024-05-10 | 朴道征信有限公司 | Data source call load balancing method, device, electronic equipment and medium |
| Publication number | Publication date |
|---|---|
| CN105959395B (en) | 2019-04-19 |
| Publication | Publication Date | Title |
|---|---|---|
| CN105959395A (en) | Cluster self-feedback type load balancing scheduling system and method | |
| CN110209494B (en) | Big data-oriented distributed task scheduling method and Hadoop cluster | |
| CN107657530A (en) | The processing method and system of a kind of operation flow | |
| CN103369054B (en) | A kind of acquisition tasks management method and system | |
| CN103491024B (en) | Job scheduling method and device for streaming data | |
| CN104915407A (en) | Resource scheduling method under Hadoop-based multi-job environment | |
| CN103927231B (en) | The energy optimization data set distribution method that a kind of data-oriented processes | |
| CN102469126B (en) | Application scheduling system, method thereof and related device | |
| CN108549583A (en) | Big data processing method, device, server and readable storage medium storing program for executing | |
| CN102307133A (en) | A method for scheduling virtual machines on a public cloud platform | |
| US20200379813A1 (en) | Allocation of cloud computing resources | |
| CN102387173A (en) | MapReduce system and method and device for scheduling tasks thereof | |
| CN110502349A (en) | An orchestration method for building cloud container applications | |
| TWI786527B (en) | User code operation method of programming platform, electronic equipment and computer-readable storage medium | |
| CN105094982A (en) | Multi-satellite remote sensing data processing system | |
| CN104601664A (en) | Cloud computing platform resource management and virtual machine dispatching control system | |
| CN106131141B (en) | A kind of distributed type assemblies load balancing parallel dispatch system and method | |
| CN112099930A (en) | Quantum computer cluster distributed queue scheduling method | |
| CN105094984A (en) | Resource scheduling method and system | |
| CN106126344A (en) | A kind of method for allocating tasks and device | |
| CN103957280A (en) | Connection allocation and scheduling method of sensor network in Internet of things | |
| CN110300188A (en) | Data transmission system, method and apparatus | |
| Wang et al. | Dependency-aware network adaptive scheduling of data-intensive parallel jobs | |
| CN117909061A (en) | Model task processing system and resource scheduling method based on GPU hybrid cluster | |
| CN104765644A (en) | Resource collaboration evolution system and method based on intelligent agent |
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20190419 Termination date:20210615 |