CN105577756A - Distributed database log collection and load regulation system adopting cross backup and method thereof - Google Patents
Distributed database log collection and load regulation system adopting cross backup and method thereof Download PDFInfo
- Publication number
- CN105577756A CN105577756A CN201510932777.6A CN201510932777A CN105577756A CN 105577756 A CN105577756 A CN 105577756A CN 201510932777 A CN201510932777 A CN 201510932777A CN 105577756 A CN105577756 A CN 105577756A
- Authority
- CN
- China
- Prior art keywords
- load
- node
- machine
- log collection
- request
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/069—Management of faults, events, alarms or notifications using logs of notifications; Post-processing of notifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a distributed database log collection and load regulation system adopting cross backup and a method thereof. According to the system, information related to the load in each machine is collected by utilizing log collection nodes, then the load situation of each machine is assessed by utilizing load regulation nodes so as to obtain a load score, and request distribution nodes give priority of the request to the corresponding machine of the lowest load pressure to perform according to the load score of each machine. The request distribution to the low-load machines can be assisted by utilizing the load information of the machines, and the high-load machines are protected as far as possible so that damage to the machines caused by the fact that the high-load machines continuously receive the request can be avoided and the short plate problem can be greatly avoided.
Description
Technical field
The present invention relates to distributed data base technical field of memory, particularly relate to the distributed data base log collection and load regulating system and method thereof that adopt and intersect and back up.
Background technology
At current cloud computing and large data age, at every field and the industry-by-industry of society, all there is the demand to mass data storage and calculating.Traditional unit storage system and unit Database Systems, facing to the pressure of mass data and impact, cannot complete storing with calculating of task at all.And the development of distributed computing technology, high speed network transmission technology and parallel computing is with ripe, distributed data base is made to become the mainstream development trend on current data storage circle.
Distributed data base great majority are the concurrent operating modes adopting many memory nodes, some data distribution algorithms is utilized to spread in each machine of database by magnanimity business datum, to expect that every platform Work machine can share overall data amount equally, make the data volume of each machine moderate like this, reduce storage and the calculating pressure of each machine.The Internet giant that current industry is well-known, the distributed memory system of such as Goolge is exactly block data being divided into fixed size, these data blocks is spread in each memory node with certain Distribution Algorithm.And the product of the well-known distributed data base manufacturer of current industry is also like this, such as Vertica and GreenPlum etc. adopt mass data to spread in each Work machine, solve the problem of mass data storage and calculating with data-base cluster.GreenPlum adopt be intersect between many memory nodes backup mode to deposit Backup Data.
Distributed data base needs to carry out multiple backup to keep high availability, high-performance and highly to expand flexibility to data fragmentation naturally.The business demand in the present age, can carry out the access of high concurrent, large pressure to distributed data base, if do not have multiple backup burst, when after main burst machine breakdown, system just externally cannot provide service, and this cannot accept.And the problem adopting the framework of many data fragmentations to bring is exactly data redundancy problem, in addition, problem of load balancing is also had.If the load of some machine is apparently higher than other machines in data-base cluster, so these machines damage possibly at short notice, cause short-board effect.
There is the mode of some simple proof load equilibriums, such as can by request with one of mode available machines used being issued to this request of the mode of RoundRobin or random selecting.But there is very large defect in these methods, namely not fully taking into account the loading condition of each machine in system, is a kind of regulative mode of blindness, and therefore effect is not fine.
The log management problem of distributed data base also needs to consider, how collecting efficiently and having managed log information is a major issue that will consider.If the loading condition of system can be obtained from log information, and utilize these load data guide data distribution module dispense request, be then a more efficient and rational solution, and any added burden can not be caused to system.
Summary of the invention
For solving the problem; realize the load regulation utilizing log information guidance system, the invention provides a kind of the distributed data base log collection and the load regulating system that adopt intersection backup, system load balancing as much as possible can be guaranteed; the machine of protection high capacity, in order to avoid cause machine breakdown.
For achieving the above object, the technical solution used in the present invention is:
A kind of distributed data base log collection and load regulating system adopting intersection backup, comprise: log collection node, for regularly obtaining log information from database each machine, and the statistical information of regarding system load in log information is sent to load regulation node; Load regulation node, for being compared with pre-configured load pressure rank by the load information of each for database machine, assesses the loading condition of each machine, is sent to Requests routing node by assessing the load score value obtained; Requests routing node, for the load score value according to each machine, request user issued preferentially sends to the machine of low load pressure to process, if the machine at the data fragmentation place that certain request is corresponding is all in the state of high pressure, then request is cached by Requests routing node, suspend the distribution of this request, until when having the loading condition of available machines used to be updated to the state of non-high pressure, then give this machine by the Requests routing of buffer memory.
Log collection mechanism is peeled off by the present invention from each Work machine, extra log collection node and unified daily record reclaim mechanism is utilized to collect all kinds of log informations of each Work machine, log information is classified, and different types of log information is sent to different processing nodes, and for the log information relevant with system load, load regulation node is then sent to carry out statistics and the quantification of information, the load information of quantification sends to Requests routing node to carry out load balancing control by load regulation node, the load balancing ensureing each burst (every platform machine) is gone in the mode of coarseness.
Log collection node of the present invention is to obtain different log informations in the different time interval to each Work machine.For real-time change or change comparatively frequently log information obtain with the shorter time interval, the real time load information etc. of such as machine; And the log information of non-frequent variations will be obtained with the longer time interval, such as machine process crash info etc.For the statistical information relevant to machine loading, load regulation node to be sent it in time.
Load regulation node needs the parameter evaluation utilizing Administrator to go out the loading condition of every platform machine, the evaluation process of load regulation node is: load regulation node gets the log information of each machine loading situation from log collection node, by these information updatings in each the machine resources use information oneself safeguarded, and the loading condition of resource threshold to every platform machine of the CPU pre-set according to DBA, internal memory, IO is given a mark, namely to every platform machine give one from 1 to 5 score value, the higher machine loading of score value is higher.
The load score value of these machines is regularly sent to Requests routing node by load regulation node, and Requests routing node can utilize the score information of machine to carry out the load regulation of coarseness.Be specially: after Requests routing node receives user's request, obtain the machine group at the data fragmentation place corresponding to this request, then search the load score value of this group machine, therefrom find the machine that a load score value is minimum, request sent to this machine to go to perform.Requests routing node is provided with a cache pool, if certain load score value of available machines used corresponding to request is 5, be namely all in high-pressure state, this request can be cached in this cache pool by Requests routing node; After Requests routing node receives new machine loading score value, if find, the load score value of certain machine becomes non-5 from 5, i.e. load reduction, can find all data available bursts in the request of this machine from cache pool, and sends to this machine to go process these requests.If the number of request of the cache pool institute buffer memory of Requests routing node is greater than the threshold value that DBA is arranged, then also can send warning information to keeper.Aforesaid way can high capacity machine to greatest extent in protection system, ensures that each machine loading of database is balanced.
Log collection node, load regulation node and Requests routing node respectively comprise working node and at least one backup node, the information that working node is regularly safeguarded sends to corresponding backup node, heartbeat detection is had between its corresponding backup node of working node, if the node of the node discovery work of backup is without response, the node then backed up takes over the task of the node of this nothing response at once, and sends associated alarm information to DBA.Each node have main have standby, can when host node goes wrong, backup node can the task of adapter host node rapidly, ensures that system is unaffected.
Log collection node collector journal, log collection node send the load log information of being correlated with and send the load score value after assessment to load regulation node, load regulation node and send relevant information to the time interval in the backup node of its correspondence all by administrator configurations to the log collection node of Requests routing node and work at present, load regulation node and Requests routing node, these time intervals can regulate in real time in the process of system cloud gray model, to adapt to different business demands.
The log collection that said system adopts and load regulation method, mainly contain following some:
(1) log collection node is enabled multiple thread by certain hour interval and obtain the log information relevant to system load from every platform Work machine, deposits, and send it to load regulation node in units of machine;
(2) load regulation node utilizes the loading condition of the load information of machine loading score criteria and every platform machine to every platform machine to assess, representing the load pressure state of machine with score value form, being sent to Requests routing node by assessing the load score value obtained;
(3) Requests routing node receives the request that user issues, and finds out the collection of machines at this request corresponding data burst place, searches the load score value that in this set, every platform machine is corresponding, this request is sent to the machine of minimum load pressure to go to perform; If the machine at this request corresponding data burst place is all in high-pressure state, this request is then cached in cache pool by Requests routing node, until when having the loading condition of available machines used to be updated to the state of non-high pressure, then give this machine by the Requests routing of buffer memory.
The load information that log collection node, load regulation node, Requests routing node are regularly safeguarded backs up, and damages influential system operation to prevent the node of work at present from occurring.
The load regulation of distributed data base mechanism merges with log collection mechanism by the present invention mutually; utilize and concentrate log collection mechanism to collect out the log information relevant to the load of every platform Work machine; and these information are processed accordingly; be converted into the machine loading grade of quantification; the machine loading information utilizing these to quantize carrys out auxiliary dispense request to low loading robotics; protect the machine of high capacity as much as possible; the machine breakdown avoiding the machine of high capacity to continue the request that receives causing, can be good at the generation avoiding short slab problem.
Accompanying drawing explanation
Fig. 1 is distributed data base intersection back-up storage schematic diagram;
Fig. 2 is distributed data base log collection and load regulating system schematic diagram;
Fig. 3 is the workflow diagram that distributed data base log collection node collects load correlation log;
Fig. 4 is the workflow diagram that distributed data base load regulation node carries out load regulation;
Fig. 5 is Requests routing thread work flow chart in distributed data base Requests routing node;
Fig. 6 is that in distributed data base Requests routing node, score value upgrades thread work flow chart.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is elaborated:
Fig. 1 is the data-base cluster example adopting intersection backup mode to store data fragmentation in the present invention, 6 machines are had in this cluster, totally 6 data fragmentations, each burst all have 1 back up burst, these bursts with intersects back up mode spread in each machine of data-base cluster.
Fig. 2 is distributed data base log collection and load regulating system schematic diagram, can find out the overall architecture of system.This system comprises log collection node, load regulation node and Requests routing node three category node, three category nodes are all made up of the node worked and 1-2 backup node, carry out the transmission of heartbeat detection and relevant information between working node and backup node.Log collection node is connected with each machine of data-base cluster, and timing acquisition log information, therefrom filters out the log information relevant with machine loading.Daily record adjustment node gets the load information of each machine from log collection node, and carry out assessment marking to each machine loading situation, give a score value between 1-5, score value lower explanation machine loading is lower, otherwise represents that machine loading is higher.Requests routing node obtains from load regulation node the machine loading information that quantizes and utilizes the distribution of these information guiding requests, and request is mail in the available machines used of corresponding low load as much as possible.
Figure 3 shows that the log collection node log information that collecting robot load is relevant from cluster in the present invention, and simple process forwards the workflow of log information, concrete steps are as described below:
Step S301, judge whether to arrive and obtain the time interval T of each machine loading correlation log, if do not arrive, continue to wait for jump procedure S301 after some system clocks, if the time of advent interval T; continue to perform step S302.
Step S302, enable multiple thread from each Work machine, obtain the log information relevant to system load such as CPU usage, internal memory service condition, system IO, and these information are cached in units of different machines.
Step S303, for each machine, its all load information is left concentratedly, and these load informations are sent to load regulation node, operate for follow-up load regulation process.
Step S304, the machine loading information in step S303 is sent to the log collection node of backup, occur situation about damaging to prevent the log collection node of work at present.
Step S305, judge whether to have updated the time interval obtaining log information, if the time interval of the new acquisition daily record of Administrator, then perform step S306, otherwise jump to step S307.
Step S306, utilize Administrator new acquisition daily record the time interval upgrade T.
Step S307, complete one and take turns workflow.
Figure 4 shows that in the present invention, load regulation node receives the relevant log information of load from log collection node, the loading condition for every platform machine carries out quantizing and forwards the workflow of quantitative information, and concrete steps are as described below:
Step S401, load regulation node receive that log collection node sends through the simply dealt log information relevant to each Work machine load.
Step S402, judge whether keeper have updated the score criteria of machine loading, if do not upgrade the score criteria of machine loading, then jump to step S404, otherwise order performs step S403.The score criteria of the machine loading of acquiescence, for CPU usage, is specially:
Score value 1:CPU utilization rate is 0% ~ 20%;
Score value 2:CPU utilization rate is 20% ~ 40%;
Score value 3:CPU utilization rate is 40% ~ 60%;
Score value 4:CPU utilization rate is 60% ~ 80%;
Score value 5:CPU utilization rate is more than 80%.
Step S403, the machine loading score criteria utilizing keeper to upgrade go to upgrade the existing machine loading score criteria of the machine.
Step S404, the load utilizing the loading condition of machine loading score criteria and each machine to be each machine are given a mark, and certain value of 1 to 5 is given in the load being each machine, larger from the load pressure of this machine of the larger expression of 1 to 5 numerical value.
Step S405, the load score information of each machine is sent to Requests routing node, for follow-up load regulation process operation, to guarantee load balancing, reduce the pressure of high capacity machine.
Step S406, the machine loading score value in step S405 is sent to the load regulation node of backup, occur situation about damaging to prevent the load regulation node of work at present.
2 main worker threads are had, i.e. Requests routing thread and score value more new thread in Requests routing node.Figure 5 shows that the workflow of Requests routing thread, concrete steps are as described below:
Step S501, Requests routing thread receive the request that user issues, and find out the set M of the machine at this request corresponding data burst place.Namely gather in M and comprise some machines, this request can go to perform in any machine wherein.
Step S502, each machine in set M is removed to search corresponding load score value, judge the load score value of all machines in set M, if the load score value of all machines is not all 5, jump to step S504, otherwise order performs step S503.
When the load score value of step S503, all machines is all 5, request is cached in the cache pool in Requests routing machine by Requests routing thread, when waiting for that the load score value of in its available machines used a certain becomes non-5 from 5, by score value more new thread go to distribute and perform these requests in cache pool.Finally, step S506 is jumped to.
Step S504, Requests routing thread find the machine that in set M, load score value is minimum, are designated as P.If there is the machine that multiple stage load score value is minimum, then chooses first and be designated as P.
This request sends to machine P to go to perform by step S505, Requests routing thread.
Step S506, complete one and take turns workflow.
Figure 6 shows that the workflow of the more new thread of score value in Requests routing node, concrete steps are as described below:
Step S601, score value more new thread timing from load regulation reception at Node to the load score information of each up-to-date machine.
Step S602, score value more new thread utilize the load score information received to remove the load score value of each the machine upgrading self maintained.In addition score value more new thread record score value becomes non-5 all machines from 5, be designated as set N.
Step S603, score value more new thread search the request whether having and wait for machine in set N in cache pool, if do not find such request, then jump to step S605, otherwise order perform step S604.
Step S604, score value more new thread are postponed to deposit in pond those are waited for please the seeking out of machine in set N, and the machine be sent in corresponding set N goes to perform.
Step S605, complete one and take turns process.
Last it is noted that above each embodiment is only in order to illustrate technical scheme of the present invention, be not intended to limit; Although with reference to foregoing embodiments to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein some or all of technical characteristic; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.
Claims (9)
1. one kind adopts the distributed data base log collection and load regulating system of intersecting and backing up, it is characterized in that comprising: log collection node, for regularly obtaining log information from database each machine, and the statistical information of regarding system load in log information is sent to load regulation node;
Load regulation node, for being compared with pre-configured load pressure rank by the load information of each for database machine, assesses the loading condition of each machine, is sent to Requests routing node by assessing the load score value obtained;
Requests routing node, for the load score value according to each machine, request user issued preferentially sends to the machine of low load pressure to process, if the machine at the data fragmentation place that certain request is corresponding is all in the state of high pressure, then request is cached by Requests routing node, suspend the distribution of this request, until when having the loading condition of available machines used to be updated to the state of non-high pressure, then give this machine by the Requests routing of buffer memory.
2. distributed data base log collection and the load regulating system adopting intersection backup according to claim 1, is characterized in that: log collection node is to obtain different log informations in the different time interval to each Work machine.
3. distributed data base log collection and the load regulating system adopting intersection backup according to claim 1, it is characterized in that: the evaluation process of load regulation node is: load regulation node gets the log information of each machine loading situation from log collection node, by these information updatings in each the machine resources use information oneself safeguarded, and according to the CPU that DBA pre-sets, internal memory, the loading condition of resource threshold to every platform machine of IO is given a mark, namely to every platform machine give one from 1 to 5 score value, the higher machine loading of score value is higher.
4. distributed data base log collection and the load regulating system adopting intersection backup according to claim 3, it is characterized in that: after Requests routing node receives user's request, obtain the machine group at the data fragmentation place corresponding to this request, then the load score value of this group machine is searched, therefrom find the machine that a load score value is minimum, request sent to this machine to go to perform.
5. distributed data base log collection and the load regulating system adopting intersection backup according to claim 4, it is characterized in that: Requests routing node is provided with a cache pool, if certain load score value of available machines used corresponding to request is 5, namely be all in high-pressure state, this request can be cached in this cache pool by Requests routing node; After Requests routing node receives new machine loading score value, if find, the load score value of certain machine becomes non-5 from 5, i.e. load reduction, can find all data available bursts in the request of this machine from cache pool, and sends to this machine to go process these requests.
6. the distributed data base log collection of the employing intersection backup according to any one of claim 1 to 5 and load regulating system, it is characterized in that: log collection node, load regulation node and Requests routing node respectively comprise working node and at least one backup node, the information that working node is regularly safeguarded sends to corresponding backup node, heartbeat detection is had between its corresponding backup node of working node, if the node of the node discovery work of backup is without response, the node then backed up takes over the task of the node of this nothing response at once, and send associated alarm information to DBA.
7. distributed data base log collection and the load regulating system adopting intersection backup according to claim 6, it is characterized in that: log collection node collector journal, log collection node sends the relevant log information of load to load regulation node, load regulation node sends the log collection node of the load score value after assessment to Requests routing node and work at present, load regulation node and Requests routing node send relevant information to the time interval in the backup node of its correspondence all by administrator configurations, these time intervals can regulate in real time in the process of system cloud gray model.
8. the log collection of the distributed data base log collection that the employing intersection according to aforementioned any one claim backs up and load regulating system and load regulation method, is characterized in that:
(1) log collection node is enabled multiple thread by certain hour interval and obtain the log information relevant to system load from every platform Work machine, deposits, and send it to load regulation node in units of machine;
(2) load regulation node utilizes the loading condition of the load information of machine loading score criteria and every platform machine to every platform machine to assess, representing the load pressure state of machine with score value form, being sent to Requests routing node by assessing the load score value obtained;
(3) Requests routing node receives the request that user issues, and finds out the collection of machines at this request corresponding data burst place, searches the load score value that in this set, every platform machine is corresponding, this request is sent to the machine of minimum load pressure to go to perform; If the machine at this request corresponding data burst place is all in high-pressure state, this request is then cached in cache pool by Requests routing node, until when having the loading condition of available machines used to be updated to the state of non-high pressure, then give this machine by the Requests routing of buffer memory.
9. log collection according to claim 8 and load regulation method, is characterized in that: the load information that log collection node, load regulation node, Requests routing node are regularly safeguarded backs up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510932777.6A CN105577756A (en) | 2015-12-12 | 2015-12-12 | Distributed database log collection and load regulation system adopting cross backup and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510932777.6A CN105577756A (en) | 2015-12-12 | 2015-12-12 | Distributed database log collection and load regulation system adopting cross backup and method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105577756A true CN105577756A (en) | 2016-05-11 |
Family
ID=55887400
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510932777.6A Pending CN105577756A (en) | 2015-12-12 | 2015-12-12 | Distributed database log collection and load regulation system adopting cross backup and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105577756A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107547429A (en) * | 2017-07-28 | 2018-01-05 | 新华三信息安全技术有限公司 | One kind load determines method, apparatus and electronic equipment |
| CN108737573A (en) * | 2018-06-28 | 2018-11-02 | 郑州云海信息技术有限公司 | A kind of distributed storage cluster and its service response control method, device and equipment |
| CN109086292A (en) * | 2018-06-11 | 2018-12-25 | 平安科技（深圳）有限公司 | A kind of switching method and system of database |
| CN111897816A (en) * | 2020-07-16 | 2020-11-06 | 中国科学院上海微系统与信息技术研究所 | Interactive method for computing information between satellites and generation method of information table applied by interactive method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090287744A1 (en) * | 2008-05-15 | 2009-11-19 | International Business Machines Corporation | Apparatus, system and method for healthcheck of information technology infrastructure based on log data |
| CN101916298A (en) * | 2010-08-31 | 2010-12-15 | 深圳市赫迪威信息技术有限公司 | Database operation method, apparatus and system |
| US20120226918A1 (en) * | 2011-03-02 | 2012-09-06 | Rallo Aaron J | Non-intrusive Power Management |
| CN104376087A (en) * | 2014-11-19 | 2015-02-25 | 天津南大通用数据技术股份有限公司 | Load balance calculation method for distributed database adopting cross backups |
-
2015
- 2015-12-12 CN CN201510932777.6A patent/CN105577756A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090287744A1 (en) * | 2008-05-15 | 2009-11-19 | International Business Machines Corporation | Apparatus, system and method for healthcheck of information technology infrastructure based on log data |
| CN101916298A (en) * | 2010-08-31 | 2010-12-15 | 深圳市赫迪威信息技术有限公司 | Database operation method, apparatus and system |
| US20120226918A1 (en) * | 2011-03-02 | 2012-09-06 | Rallo Aaron J | Non-intrusive Power Management |
| CN104376087A (en) * | 2014-11-19 | 2015-02-25 | 天津南大通用数据技术股份有限公司 | Load balance calculation method for distributed database adopting cross backups |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107547429A (en) * | 2017-07-28 | 2018-01-05 | 新华三信息安全技术有限公司 | One kind load determines method, apparatus and electronic equipment |
| CN109086292A (en) * | 2018-06-11 | 2018-12-25 | 平安科技（深圳）有限公司 | A kind of switching method and system of database |
| CN109086292B (en) * | 2018-06-11 | 2023-11-28 | 平安科技（深圳）有限公司 | Database switching method and system |
| CN108737573A (en) * | 2018-06-28 | 2018-11-02 | 郑州云海信息技术有限公司 | A kind of distributed storage cluster and its service response control method, device and equipment |
| CN111897816A (en) * | 2020-07-16 | 2020-11-06 | 中国科学院上海微系统与信息技术研究所 | Interactive method for computing information between satellites and generation method of information table applied by interactive method |
| CN111897816B (en) * | 2020-07-16 | 2024-04-02 | 中国科学院上海微系统与信息技术研究所 | Interaction method of calculation information between satellites and generation method of information table applied by same |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9128899B1 (en) | Predictive failover planning | |
| CN104391737B (en) | The optimization method of load balance in cloud platform | |
| CN104462225B (en) | The method, apparatus and system of a kind of digital independent | |
| US10298715B2 (en) | Distributed processing system, task processing method, and storage medium | |
| CN102694868A (en) | Cluster system implementation and task dynamic distribution method | |
| CN105577756A (en) | Distributed database log collection and load regulation system adopting cross backup and method thereof | |
| CN103558992A (en) | Off-heap direct-memory data stores, methods of creating and/or managing off-heap direct-memory data stores, and/or systems including off-heap direct-memory data store | |
| US20180357727A1 (en) | Methods and apparatuses for adjusting the distribution of partitioned data | |
| CN103744906A (en) | System, method and device for data synchronization | |
| CN104504147B (en) | A kind of resource coordination method of data-base cluster, apparatus and system | |
| CN101137984A (en) | Systems, methods, and software for distributed loading of databases | |
| CN104462432A (en) | Self-adaptive distributed computing method | |
| CN103746838A (en) | Task scheduling method of computer network without center node | |
| US8843620B2 (en) | Monitoring connections | |
| CN110727508A (en) | Task scheduling system and scheduling method | |
| CN109154933B (en) | Distributed database system and method for distributing and accessing data | |
| CN111459642A (en) | Fault processing and task processing method and device in distributed system | |
| JP2014228983A (en) | Work management apparatus, and work management method, and computer program | |
| CN106201772A (en) | The backup of a kind of operating system based on data center, restoration methods and device | |
| CN115168042A (en) | Management method and device of monitoring cluster, computer storage medium and electronic equipment | |
| US11134121B2 (en) | Method and system for recovering data in distributed computing system | |
| CN111078494B (en) | Cache database monitoring method and server | |
| CN112035262B (en) | Method and device for dynamically managing and adjusting multiple hosts | |
| EP2828761A1 (en) | A method and system for distributed computing of jobs | |
| EP3709173A1 (en) | Distributed information memory system, method, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160511 |