技术领域technical field
本发明具体地说是一种改进的HBASE负载均衡策略。Specifically, the invention is an improved HBASE load balancing strategy.
背景技术Background technique
随着社交应用平台及电子商务类互联网应用的不断出现，标志着人们已经步入更高级的生活时代。在这种海量数据、高并发访问的情况下，数据库能提供稳定、可靠的性能是至关重要的。但传统的数据库已经不能满足这种大数据的问题，在这种技术背景下，分布式的数据库应运而生。HBase是Hadoop上的一个NoSQL数据库的实现，同时也是一个分布式的、面向列的开源数据库。它提供了强大的可扩展性和数据操作性能。在HBase中，Master 负责分配 Region 到各个不同的 Region Server 上，Region Server 负责存储实际的数据。但如果Region Server张的负载不均衡或者每个Region Server中的服务请求数量不均衡，这些都将影响HBase的服务性能。因此，HBase负载均衡策略是非常重要的。 With the continuous emergence of social application platforms and e-commerce Internet applications, it marks that people have entered a more advanced life era. In the case of massive data and high concurrent access, it is very important for the database to provide stable and reliable performance. However, traditional databases can no longer meet the problems of such big data. Under this technical background, distributed databases emerge as the times require. HBase is an implementation of a NoSQL database on Hadoop, and it is also a distributed, column-oriented open source database. It provides strong scalability and data manipulation performance. In HBase, the Master is responsible for assigning Regions to different Region Servers, and the Region Servers are responsible for storing actual data. However, if the load of the Region Servers is unbalanced or the number of service requests in each Region Server is unbalanced, these will affect the service performance of HBase. Therefore, the HBase load balancing strategy is very important. the
目前默认的HBase负载均衡算法是单纯的均衡Region Server中的Region数量。该算法中首先将所有Region Server以含有Region的数量进行排序，如果负载相同则按照ServerName进行排序。然后根据默认算法判断是否需要进行负载均衡，若需要负载均衡则计算出最小负载和最大负载，将每个Region Server中超出最大负载值的剩下的Region记录在regionsToMove中。再次，将小于最小负载的Region Server从regionsToMove队列中取出全部加到最小负载中。最后查看所有小于最小负载Region Server都升到最小负载时， regionsToMove队列还有剩余region的时候则将遍历最小负载的Region Server起始达到最大值。若是还有RS负载小于最小值的时候从最大负载Region Server中取得需要的region数放入regionsToMove队列，在进行region分配，使得小于最小值的Region Server达到最小负载。但是该方法会导致过热数据分布不均匀的情况，因此，将每个Regoin的最近服务请求数作为平衡过热数据的指标，综合平衡Region Server中Region数量及平衡过热数据这两个因素提出一种改进的HBase负载均衡策略。该策略可以使得每台Region Server上的Region数量及服务的请求数量相对均衡，改进过热数据访问的情况，提高服务请求的访问性能，同时达到负载均衡的目的。The current default HBase load balancing algorithm is to simply balance the number of Regions in the Region Server. In this algorithm, all Region Servers are first sorted by the number of Regions, and if the loads are the same, they are sorted by ServerName. Then judge whether load balancing is required according to the default algorithm. If load balancing is required, calculate the minimum load and maximum load, and record the remaining Regions exceeding the maximum load value in each Region Server in regionsToMove. Again, all the Region Servers that are smaller than the minimum load are taken from the regionsToMove queue and added to the minimum load. Finally, when all the Region Servers that are less than the minimum load have risen to the minimum load, when there are still regions left in the regionsToMove queue, it will traverse the Region Server with the minimum load to reach the maximum value. If there are still RS loads that are less than the minimum value, the required number of regions is obtained from the maximum load Region Server and put into the regionsToMove queue, and the region distribution is performed so that the Region Servers that are less than the minimum value reach the minimum load. However, this method will lead to uneven distribution of overheating data. Therefore, the number of recent service requests for each Regoin is used as an indicator for balancing overheating data, and an improvement is proposed to comprehensively balance the two factors of the number of Regions in the Region Server and the balance of overheating data. HBase load balancing strategy. This strategy can make the number of Regions on each Region Server and the number of service requests relatively balanced, improve the situation of overheating data access, improve the access performance of service requests, and achieve the purpose of load balancing at the same time.
发明内容Contents of the invention
本发明的目的是克服现有技术中存在的不足，提供一种改进的HBASE负载均衡策略，针对平衡Region Server中Region数量及每个Region的最近服务请求数平衡过热数据这两个因素作为负载均衡的依据，从而实现负载均衡的目的，同时使得每台Region Server上的Region数量及服务的请求数量相对均衡。The purpose of the present invention is to overcome the deficiencies in the prior art, to provide an improved HBASE load balancing strategy, aiming at balancing the two factors of the number of Regions in the Region Server and the number of recent service requests of each Region to balance overheating data as load balancing In order to achieve the purpose of load balancing, the number of Regions and the number of service requests on each Region Server are relatively balanced.
本发明的技术方案是按以下方式实现的，其结构中首先根据每个Region Server中Region的数量计算出每个Region Server中是否需要Region迁移，并确定Region Server中需要迁移的Region个数；然后，根据每个Region Server中的Region最近服务请求数决定Region Server中迁移的具体的Region；综合上述两个方面可以达到每个Region Server中Region个数均衡同时也均衡了过热数据访问的情形；The technical solution of the present invention is realized in the following manner, in its structure, at first calculate whether Region migration is needed in each Region Server according to the quantity of Regions in each Region Server, and determine the number of Regions that need to be migrated in the Region Server; then According to the number of recent service requests in each Region Server, the specific Region to be migrated in the Region Server is determined; combining the above two aspects can achieve a balance in the number of Regions in each Region Server and also balance the situation of overheating data access;
确定Region Server中需要迁移的Region数量的模块中是根据Region的数量计算出最大最小负载值，并根据最大最小负载值确定Region Server中需要迁移的Region个数；首先将所有Region Server以含有Region的数量进行排序，如果负载相同则按照ServerName进行排序；然后根据默认算法判断是否需要进行负载均衡；再次，若需要负载均衡则计算出最小负载和最大负载，其中最小负载等于Region总数与Region Server总数之比；最大负载是根据Region总数与Region Server总数取余，若为0 ，则最大负载等于最小值，否则最大负载等于最大值；最后根据最大负载值为上限，将大于最大负载值得Region Server中的Region个数记录到regionsToMove中；In the module for determining the number of Regions to be migrated in the Region Server, the maximum and minimum load values are calculated according to the number of Regions, and the number of Regions to be migrated in the Region Server is determined according to the maximum and minimum load values; If the load is the same, it will be sorted according to the ServerName; then judge whether load balancing is required according to the default algorithm; again, if load balancing is required, calculate the minimum load and maximum load, where the minimum load is equal to the difference between the total number of Regions and the total number of Region Servers Ratio; the maximum load is based on the remainder of the total number of Regions and the total number of Region Servers. If it is 0, the maximum load is equal to the minimum value, otherwise the maximum load is equal to the maximum value; finally, according to the upper limit of the maximum load value, it will be greater than the maximum load value in the Region Server The number of Regions is recorded in regionsToMove;
针对Region最近所服务的请求数平衡过热数据访问模块首先根据每个Region最近的服务请求数计算出最小请求数和最大请求数，然后根据Region Server中服务请求数的多少决定选择需要迁移的Region；最小请求数是根据每个Region的总请求数与Region Server总数之比；最大请求数是Region的总请求数与Region Server总数取余，若为0，则最大请求数等于最小请求数与最大单个Region请求数之间取最大值，否则最大请求数等于最小请求书加上最小单个Region请求数与最大单个Region请求数之间取最大值；即最大请求数=max{Region总请求数%RegionServer总数==0？最小请求数：最小请求数+最小单个Region请求数，最大单个Region请求数}；然后根据Region Server中需要迁移的Region个数选择需要迁移的Region，此时Region的选择是根据最大请求数决定的，将超过最大请求数的值记录下来M，然后将Region Server中最接近M的Region迁移出去；放在需要Region数量的且需要服务请求数的Region Server中。The data access module first calculates the minimum number of requests and the maximum number of requests based on the number of recent service requests in each Region, and then selects the Region to be migrated according to the number of service requests in the Region Server; The minimum number of requests is based on the ratio of the total number of requests in each Region to the total number of Region Servers; the maximum number of requests is the remainder of the total number of requests in a Region and the total number of Region Servers. If it is 0, the maximum number of requests is equal to the minimum number of requests and the largest single The maximum value is taken between the number of Region requests, otherwise the maximum number of requests is equal to the minimum request book plus the maximum value between the minimum number of single Region requests and the maximum number of single Region requests; that is, the maximum number of requests = max{Total number of Region requests% Total number of RegionServers ==0? Minimum number of requests: minimum number of requests + minimum number of single Region requests, maximum number of single Region requests}; then select the Region to be migrated according to the number of Regions that need to be migrated in the Region Server. At this time, the selection of the Region is determined by the maximum number of requests , record the value M that exceeds the maximum number of requests, and then migrate the Region that is closest to M in the Region Server; put it in the Region Server that requires the number of Regions and the number of service requests.
本发明的优点是：The advantages of the present invention are:
本发明的一种改进的HBASE负载均衡策略和现有技术相比，有效根据集群中Region Server中Region数量及Region的服务请求数量这两个因素作为HBase负载均衡的依据来达到负载均衡的目的；此方法带来的效果首先是保证了每个Region Server中的Region个数的均衡，在数据访问时增加并行度；其次是在此基础上改进了过热数据访问的情形，提高服务请求的访问性能。Compared with the prior art, an improved HBASE load balancing strategy of the present invention effectively achieves the purpose of load balancing according to the two factors of the Region quantity in the Region Server in the cluster and the service request quantity of the Region as the basis for HBase load balancing; The effect of this method is first to ensure the balance of the number of Regions in each Region Server, and to increase the degree of parallelism in data access; secondly, to improve the situation of overheated data access on this basis, and to improve the access performance of service requests .
附图说明Description of drawings
图1为HBase负载均衡策略流程图。Figure 1 is a flow chart of the HBase load balancing strategy.
图2集群中每个Region Server中Region数量。Figure 2 shows the number of Regions in each Region Server in the cluster.
图3欲迁移Region Server中Region的数量。Figure 3 shows the number of Regions in the Region Server to be migrated.
图4 Region Server中的Region最近请求数量。Figure 4 The number of recent Region requests in the Region Server.
实施方式Implementation
下面结合附图对本发明的一种改进的HBASE负载均衡策略作以下详细说明。An improved HBASE load balancing strategy of the present invention will be described in detail below in conjunction with the accompanying drawings.
如图1所示，本发明的一种改进的HBASE负载均衡策略针对分布式集群中每个Region Server上的Region数量进行均衡及根据Region最近所服务的请求数平衡过热数据访问，把这两个因素有效的结合起来作为负载均衡的依据，其优势在于使得每台Region Server上的Region数量及服务的请求数量相对均衡，改进过热数据访问的情况，提高服务请求的访问性能，同时达到负载均衡的目的。As shown in Figure 1, an improved HBASE load balancing strategy of the present invention balances the number of Regions on each Region Server in the distributed cluster and balances overheating data access according to the number of requests served by the Region recently. Factors are effectively combined as the basis for load balancing. Its advantage is to make the number of Regions on each Region Server and the number of service requests relatively balanced, improve the situation of overheating data access, improve the access performance of service requests, and achieve load balancing. Purpose.
均衡每个Region Server上的Region数量是根据默认HBase的负载均衡算法得到需要迁移的Region数量，使得每个Region Server上的Region数量相对均衡。该算法中首先将所有Region Server以含有Region的数量进行排序，如果负载相同则按照ServerName进行排序。然后根据默认算法判断是否需要进行负载均衡。再次，若需要负载均衡则计算出最小负载和最大负载，最后将需要迁移的Region Server中的Region数量记录在regionsToMove中。Balancing the number of Regions on each Region Server is based on the default HBase load balancing algorithm to obtain the number of Regions that need to be migrated, so that the number of Regions on each Region Server is relatively balanced. In this algorithm, all Region Servers are first sorted by the number of Regions, and if the loads are the same, they are sorted by ServerName. Then judge whether load balancing needs to be performed according to the default algorithm. Again, if load balancing is required, calculate the minimum load and maximum load, and finally record the number of Regions in the Region Server to be migrated in regionsToMove.
针对Region最近所服务的请求数平衡过热数据访问模块是根据服务请求数量的平均值及最大值最小值策略计算出每个Region Server中有多少请求数量需要迁移。根据需要迁移的请求数量选择regionsToMove中每个Region Sever需要迁移的特定Region。特定Region的选择是按照Region Server中需要迁移的请求数及需要迁移Region的个数进行综合，找到满足需要迁移的Region数量同时这些Region最近服务的请求数之和与该Region Server中需要迁移的服务请求数最接近。将满足这两个因素的特定Region迁移到需要Region个数均衡的且缺少服务请求数的Region Server中。The data access module calculates how many requests need to be migrated in each Region Server based on the average number of service requests and the maximum and minimum value strategy. Select the specific Region to be migrated for each Region Sever in regionsToMove according to the number of requests to be migrated. The selection of a specific Region is based on the number of requests that need to be migrated in the Region Server and the number of Regions that need to be migrated, and the number of Regions that meet the needs of migration is found. At the same time, the sum of the number of recent service requests of these Regions and the services that need to be migrated in the Region Server The number of requests is closest. Migrate a specific Region that meets these two factors to a Region Server that requires a balanced number of Regions and lacks the number of service requests.
均衡每个Region Server上的Region数量是指满足Region Server负载均衡的能力。它是根据默认的负载均衡算法得到Region Server的迁移数量，该模块只负责确定Region Server中需要迁移的Region个数。具体的迁移哪个Region则是根据新增模块，即平衡过热数据模块来决定的。 Balancing the number of Regions on each Region Server refers to the ability to satisfy the load balancing of Region Servers. It obtains the number of Region Server migrations based on the default load balancing algorithm, and this module is only responsible for determining the number of Regions that need to be migrated in the Region Server. The specific region to migrate is determined based on the newly added module, that is, the balance overheating data module. the
针对Region最近所服务的请求数平衡过热数据访问模块主要考虑数据过热问题。根据此模块可以综合考虑Region Server中服务请求情况，改进过热数据访问的情况，提高服务请求的访问性能。The overheating data access module mainly considers the problem of data overheating based on the recent number of requests served by the Region. According to this module, the service request situation in Region Server can be considered comprehensively, the situation of overheating data access can be improved, and the access performance of service request can be improved.
具体实施方式Detailed ways
首先部署分布式集群环境，在操作系统是centos6.3上按照官方文档安装hadoop组件。然后将hdfs、mapreduce、hbase服务开启。在集群中有6个Datanode节点，每个节点上有1个Region Server。改进的HBase负载均衡策略如图1所示( RS:Region Server， R:Region)，当要进行HBase负载均衡时，首先根据Region的数量计算出最大最小负载值即最大最小R数，并根据最大最小R数值确定Region Server中需要迁移的Region个数。如果集群中RS中的R数量如图2所示：RS1是22，RS2是20，RS3是18，RS4是20，RS5是10，RS6是4。将RS中的R数量排序，即为RS6 4，RS5 10，RS3 18，RS2 20，RS4 20，RS1 22。最小R数=R总数/RS总数=15；最大R数=R总数%RS总数==0？最小R数：最小R数+1，因此最大R数=16。按照图1所示的将大于最大R数的RS中的R数量放入到regionToMove中，这时每个RS情况是：RS6 4，RS5 10，RS3 16，RS216，RS4 16，RS1 16；regionToMove中的欲迁移RS中R的数量如图3所示：RS3 2，RS2 4，RS4 4，RS1 6。First, deploy a distributed cluster environment, and install hadoop components on the operating system centos6.3 according to official documents. Then enable hdfs, mapreduce, and hbase services. There are 6 Datanode nodes in the cluster, and there is 1 Region Server on each node. The improved HBase load balancing strategy is shown in Figure 1 (RS:Region Server, R:Region). When HBase load balancing is to be performed, the maximum and minimum load values, that is, the maximum and minimum R numbers, are first calculated according to the number of Regions, and based on the maximum The minimum R value determines the number of Regions that need to be migrated in the Region Server. If the number of R in RS in the cluster is as shown in Figure 2: RS1 is 22, RS2 is 20, RS3 is 18, RS4 is 20, RS5 is 10, and RS6 is 4. Sort the number of R in RS, that is, RS6 4, RS5 10, RS3 18, RS2 20, RS4 20, RS1 22. Minimum R number = total R number / total RS = 15; maximum number of R = total number of R % total number of RS == 0? Minimum R number: Minimum R number + 1, so maximum R number = 16. Put the number of R in the RS that is greater than the maximum number of R into regionToMove as shown in Figure 1. At this time, the situation of each RS is: RS6 4, RS5 10, RS3 16, RS216, RS4 16, RS1 16; in regionToMove The number of R in the RS to be migrated is shown in Figure 3: RS3 2, RS2 4, RS4 4, RS1 6.
Region最近所服务的请求数平衡过热数据访问模块是根据regionToMove中的迁移R数量确定具体迁移的R，此确定迁移R的规则是将每个RS中欲迁移R数量的最近所服务的请求数相加，如果最近M(超过最大请求数的部分)时，则选择该组R为此RS中需要迁移的Region。在此情况下，如果集群中RS中的R最近请求数量如图4所示：RS1是330，RS2是330，RS3是270，RS4是300，RS5是150，RS6是60。最小请求数= R服务请求总数/RS总数=235；最大请求数==max{Region总请求数%RegionServer总数==0？最小请求数：最小请求数+最小单个Region请求数，最大单个Region请求数}=235，其中最小单个Region服务请求数量是0，最大单个Region服务请求数量是100。因此，需要计算迁移的RS中的R的请求数之和(根据每个需要迁移的RS的R个数)，如RS3中有两个R需要迁移，确定具体的两个R的方法是将遍历RS3中的R，使得每两个R最近请求数进行求和，选取最接近M的，在RS3的M=RS3总请求数-最大请求数=270-235=35。然后将确定下来的两个R放入MoveUpdate中，如果需要迁移RS中的R最近请求数小于最小请求数时，就随机的选取相应个数的R放入MoveUpdate中。最后根据小于最小R数的RS从MoveUpdate队列中取出加到最小R数的RS中，此过程保证此时的最近服务请求数符合最小服务请求数，如果该小于最小R数的RS中服务请求数超过最小服务请求数，则可以选取服务请求数最小的满足条件的R即可。 The number of recently served requests in the region balances the overheating data access module to determine the specific migration R according to the number of migration R in regionToMove. Plus, if the latest M (the part exceeding the maximum number of requests), select this group R as the Region that needs to be migrated in this RS. In this case, if the number of R recent requests in RS in the cluster is as shown in Figure 4: RS1 is 330, RS2 is 330, RS3 is 270, RS4 is 300, RS5 is 150, and RS6 is 60. Minimum number of requests = total number of R service requests/total number of RSs=235; maximum number of requests==max{total number of Region requests% total number of RegionServers==0? Minimum number of requests: minimum number of requests + minimum number of single Region requests, maximum number of single Region requests}=235, where the minimum number of single Region service requests is 0, and the maximum number of single Region service requests is 100. Therefore, it is necessary to calculate the sum of the requests for R in the migrated RS (according to the number of R in each RS that needs to be migrated). For example, there are two R in RS3 that need to be migrated. R in RS3 makes the sum of the most recent requests for every two Rs, and selects the one closest to M. In RS3, M=RS3 total requests-maximum requests=270-235=35. Then put the determined two Rs into MoveUpdate. If the number of recent requests for R in the migration RS is less than the minimum number of requests, randomly select the corresponding number of Rs and put them into MoveUpdate. Finally, according to the RS that is less than the minimum R number, it is taken from the MoveUpdate queue and added to the RS with the minimum R number. This process ensures that the latest service requests at this time meet the minimum service request number. If the number of service requests in the RS that is less than the minimum R number If the minimum number of service requests is exceeded, the R that satisfies the condition with the smallest number of service requests can be selected. the
据此，此方法带来的效果是首先是保证了每个Region Server中的Region个数的均衡，在此基础上改进了过热数据访问的情形，提高服务请求的访问性能，同时达到了每台Region Server上的Region数量及服务的请求数量相对均衡。Accordingly, the effect of this method is first to ensure the balance of the number of Regions in each Region Server, and on this basis to improve the situation of overheating data access, improve the access performance of service requests, and achieve the The number of Regions on the Region Server and the number of service requests are relatively balanced.
本发明的一种改进的HBASE负载均衡策略其加工制作非常简单方便，按照说明书附图所示即可加工。An improved HBASE load balancing strategy of the present invention is very simple and convenient to manufacture, and can be processed as shown in the accompanying drawings.
除说明书所述的技术特征外，均为本专业技术人员的已知技术。Except for the technical features described in the instructions, all are known technologies by those skilled in the art.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410551035.4ACN104243621A (en) | 2014-10-17 | 2014-10-17 | Improved HBASE load balancing strategy |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410551035.4ACN104243621A (en) | 2014-10-17 | 2014-10-17 | Improved HBASE load balancing strategy |
| Publication Number | Publication Date |
|---|---|
| CN104243621Atrue CN104243621A (en) | 2014-12-24 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410551035.4APendingCN104243621A (en) | 2014-10-17 | 2014-10-17 | Improved HBASE load balancing strategy |
| Country | Link |
|---|---|
| CN (1) | CN104243621A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104951561A (en)* | 2015-07-08 | 2015-09-30 | 携程计算机技术（上海）有限公司 | HBase hotspot eliminating method and system |
| CN105187498A (en)* | 2015-08-10 | 2015-12-23 | 携程计算机技术（上海）有限公司 | Region allocation method and system for HBase table |
| CN105630945A (en)* | 2015-12-23 | 2016-06-01 | 浪潮集团有限公司 | HBase region data overheating-based balancing method |
| CN107491448A (en)* | 2016-06-12 | 2017-12-19 | 中国移动通信集团四川有限公司 | A kind of HBase resource adjusting methods and device |
| CN107783720A (en)* | 2016-08-24 | 2018-03-09 | 深圳市深信服电子科技有限公司 | A kind of data balancing method and storage device |
| CN107894942A (en)* | 2017-12-04 | 2018-04-10 | 北京小度信息科技有限公司 | The monitoring method and device of tables of data visit capacity |
| CN107895030A (en)* | 2017-11-21 | 2018-04-10 | 上海帝联网络科技有限公司 | HBase Region merging methods and device, computer-readable recording medium |
| CN110019528A (en)* | 2017-12-26 | 2019-07-16 | 中国移动通信集团湖北有限公司 | Database manipulation load-balancing method, device, equipment and medium |
| CN110515726A (en)* | 2019-08-14 | 2019-11-29 | 苏州浪潮智能科技有限公司 | A database load balancing method and device |
| CN110888919A (en)* | 2019-12-04 | 2020-03-17 | 阳光电源股份有限公司 | HBase-based big data statistical analysis method and device |
| CN112395318A (en)* | 2020-11-24 | 2021-02-23 | 福州大学 | Distributed storage middleware based on HBase + Redis |
| CN112988703A (en)* | 2019-12-18 | 2021-06-18 | 中国移动通信集团四川有限公司 | Read-write request balancing method and device |
| CN116069594A (en)* | 2023-03-07 | 2023-05-05 | 武汉工程大学 | Load balancing prediction method, device and system and storage medium |
| WO2025082542A1 (en)* | 2023-10-19 | 2025-04-24 | 天翼电子商务有限公司 | Region server balancing processing method and apparatus, and storage medium and computer device |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104951561A (en)* | 2015-07-08 | 2015-09-30 | 携程计算机技术（上海）有限公司 | HBase hotspot eliminating method and system |
| CN104951561B (en)* | 2015-07-08 | 2018-09-07 | 上海携程商务有限公司 | HBase hot spots removing method and system |
| CN105187498B (en)* | 2015-08-10 | 2018-05-08 | 携程计算机技术（上海）有限公司 | The Region distribution methods and system of HBase table |
| CN105187498A (en)* | 2015-08-10 | 2015-12-23 | 携程计算机技术（上海）有限公司 | Region allocation method and system for HBase table |
| CN105630945A (en)* | 2015-12-23 | 2016-06-01 | 浪潮集团有限公司 | HBase region data overheating-based balancing method |
| CN107491448A (en)* | 2016-06-12 | 2017-12-19 | 中国移动通信集团四川有限公司 | A kind of HBase resource adjusting methods and device |
| CN107783720A (en)* | 2016-08-24 | 2018-03-09 | 深圳市深信服电子科技有限公司 | A kind of data balancing method and storage device |
| CN107895030A (en)* | 2017-11-21 | 2018-04-10 | 上海帝联网络科技有限公司 | HBase Region merging methods and device, computer-readable recording medium |
| CN107895030B (en)* | 2017-11-21 | 2020-08-14 | 上海帝联网络科技有限公司 | Region merging method and device of HBase and computer readable storage medium |
| CN107894942A (en)* | 2017-12-04 | 2018-04-10 | 北京小度信息科技有限公司 | The monitoring method and device of tables of data visit capacity |
| CN107894942B (en)* | 2017-12-04 | 2020-06-02 | 北京星选科技有限公司 | Method and device for monitoring data table access amount |
| CN110019528A (en)* | 2017-12-26 | 2019-07-16 | 中国移动通信集团湖北有限公司 | Database manipulation load-balancing method, device, equipment and medium |
| CN110515726A (en)* | 2019-08-14 | 2019-11-29 | 苏州浪潮智能科技有限公司 | A database load balancing method and device |
| CN110888919A (en)* | 2019-12-04 | 2020-03-17 | 阳光电源股份有限公司 | HBase-based big data statistical analysis method and device |
| CN112988703A (en)* | 2019-12-18 | 2021-06-18 | 中国移动通信集团四川有限公司 | Read-write request balancing method and device |
| CN112988703B (en)* | 2019-12-18 | 2022-09-16 | 中国移动通信集团四川有限公司 | Read-write request balancing method and device |
| CN112395318A (en)* | 2020-11-24 | 2021-02-23 | 福州大学 | Distributed storage middleware based on HBase + Redis |
| CN116069594A (en)* | 2023-03-07 | 2023-05-05 | 武汉工程大学 | Load balancing prediction method, device and system and storage medium |
| WO2025082542A1 (en)* | 2023-10-19 | 2025-04-24 | 天翼电子商务有限公司 | Region server balancing processing method and apparatus, and storage medium and computer device |
| Publication | Publication Date | Title |
|---|---|---|
| CN104243621A (en) | Improved HBASE load balancing strategy | |
| US11595260B2 (en) | Automatic placement of clients in a distributed computer system satisfying constraints | |
| US10455018B2 (en) | Distributed processing of shared content | |
| CN108475207B (en) | Joint auto-scaling of cloud applications | |
| CN103713956B (en) | Method for intelligent weighing load balance in cloud computing virtualized management environment | |
| CN104969213B (en) | Data flow for low latency data access is split | |
| JP6886964B2 (en) | Load balancing method and equipment | |
| US9647943B2 (en) | Method for providing tiered load balancing for a hosted voice-over internet protocol (VoIP) private branch exchange (PBX) | |
| US20150120931A1 (en) | Automatic remediation in a distributed computer system with multiple clusters of host computers | |
| WO2012173642A1 (en) | Decentralized management of virtualized hosts | |
| WO2012173641A1 (en) | Decentralized management of virtualized hosts | |
| US10192165B2 (en) | System and method for navigating multi-dimensional decision trees using acceptable alternate nodes | |
| WO2016040357A2 (en) | Load balancing of cloned virtual machines | |
| JP6588210B2 (en) | Network resource requirements for traffic through multistage switch networks | |
| CN111953732B (en) | Resource scheduling method and device in cloud computing system | |
| JP2012099062A (en) | Service cooperation system and information processing system | |
| CN103595805A (en) | Data placement method based on distributed cluster | |
| WO2017151510A1 (en) | A method and device for scheduling resources | |
| EP3069475A1 (en) | Streaming event data collection | |
| US10157076B2 (en) | Virtual machine migration device and method | |
| CN105426255B (en) | Data locality scheduling method of ReduceTask based on network I/O cost evaluation in Hadoop big data platform | |
| EP2940587A1 (en) | Computer, control device for computer system, and recording medium | |
| US10255306B2 (en) | Deduplicated data processing hierarchical rate control in a data deduplication system | |
| WO2016065198A1 (en) | High performance hadoop with new generation instances | |
| CN105373451B (en) | The method and device that a kind of virtual machine is placed |
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | Application publication date:20141224 | |
| WD01 | Invention patent application deemed withdrawn after publication |