CN103780497B - Extendible distributed coordination service management under a kind of cloud platform - Google Patents
Extendible distributed coordination service management under a kind of cloud platform Download PDFInfo
- Publication number
- CN103780497B CN103780497B CN201310749438.5A CN201310749438A CN103780497B CN 103780497 B CN103780497 B CN 103780497B CN 201310749438 A CN201310749438 A CN 201310749438A CN 103780497 B CN103780497 B CN 103780497B
- Authority
- CN
- China
- Prior art keywords
- node
- network topology
- distributed
- member node
- overall network
- 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.)
- Active
Links
Abstract
The invention discloses extendible distributed coordination service management under a kind of cloud platform, including: with the organizational form of the intermediate node intersection woods, all nodes of distributed type assemblies are initialized, all member node go out father and son's member node information of this node according to the overall network topology structure query from central administration node, carry out synchronizing information with heartbeat mechanism and communicate between all father and son's member node, member node judges whether to receive member condition modification information from other member node, if receiving, determining whether that this member condition modification information instruction member reduces and being also to increase, if reducing, if after member node receives the member condition modification information that member reduces, a member condition modification information is encapsulated out according to interrupting heart beating, and send it to central administration node.The present invention updates the efficiency of service and drastically will not decline in expansion process, is therefore more suitable for large-scale cloud platform and data handling system.
Description
Technical field
The invention belongs to field of cloud computer technology, more particularly, to extendible under a kind of cloud platform
Distributed coordination service management.
Background technology
Cloud computing and big data today prevailing, cloud service and Data Management Analysis are typically established at ten thousand grades
Even on the gallery of million grades of scales.In the cluster of such scale, node failure, disk are wrong
Missing the problem probabilities of occurrence such as the most even network failure higher, this gives service fault-tolerant and reliability proposes the biggest
Challenge, this usual employing coordination service is solved problem.Coordination service system be do not change former
There is under system architecture the best solution solving fault-tolerant problem, ensureing reliability services, by reliably
Coordination service realize the mechanism such as reliable management, distributed algorithm and event notice of cluster.?
In existing coordination service system, Google achieves lock service system Chubby, for for such as
The distributed system of BigTable, Percolator, GFS provide such as metadata management, naming service,
The reliability services such as failure monitoring；Yahoo！Coordination service system Zookeeper of increasing income of exploitation, passes through
It is reliable that the interface of asynchronous primitive realizes such as distributed queue, distributed lock, subscription Issuance model etc.
Service.Wherein, Zookeeper is mono-realization increased income of Chubby, it be one for large-scale point
Cloth system reliably coordinate system, it is provided that such as configuring maintenance, name Service, distributed synchronization,
The coordination service of group service.Its design object is the complicated error-prone key service of encapsulation, will be the easiest
Interface and the system of performance efficiency, function-stable be supplied to user.It is with quick Paxos algorithm
Based on use multiple server to service simultaneously simultaneously, central administration node coordinate internal consistency, from
And solve the "bottleneck" problem of Chubby.
But it is as cloud and processes the variation of scene and the continuous expansion of cloud platform scale, Yun Ping
Platform is more and more higher to the requirement of coordination service, and demand also gets more and more.And existing distributed coordination clothes
Business system mostly uses the framework of centralization to carry out the concordance of coordination data, and central administration node is undertaken
Work more than other nodes, when coordination service increases, Centroid than other nodes be easier to
Reaching bottleneck, cause whole concordance process performance to decline, its more new capability is as well as the expansion of system
Exhibition declines.It addition, this centralization coordination mode can occur of short duration because of the inefficacy of Centroid
Service disruption, the recovery time of service failure can be elongated along with increasing of node, there is expansion process
Middle reliabilty and availability problem.
As can be seen here, the dynamic expansion of coordination service plays the part of the biggest effect at large-scale cloud platform,
But also it is tradition coordination service system yet unresolved issue.Cloud computing is developed so far, the rule of cloud platform
Can mould constantly expands, and the requirement of coordination service is on the increase, therefore according to the scale of current system
Dynamically adjusting the scale of coordination service system with load is the most valuable thing, be also one very
There is the research contents of challenge.Therefore, how to realize dynamic extending, high performance and highly reliable
Coordination service is that under current cloud platform, coordination service system is badly in need of the matter of utmost importance solved.
Summary of the invention
For disadvantages described above or the Improvement requirement of prior art, the invention provides can under a kind of cloud platform
The distributed coordination service management of extension, it is intended that solve in existing coordination service system
The technical problem that the autgmentability existed is poor and management process efficiency is low.
For achieving the above object, according to one aspect of the present invention, it is provided that can expand under a kind of cloud platform
The distributed coordination service management of exhibition, comprises the following steps:
(1) at the beginning of all nodes of distributed type assemblies being carried out with the organizational form of the intermediate node intersection woods
Beginningization；
(2) all member node go out according to the overall network topology structure query from central administration node
Father and son's member node information of this node, carries out information with heartbeat mechanism between all father and son's member node
Synchronize and communicate；
(3) member node judges whether to receive member condition modification information from other member node, as
Fruit receives, and determines whether that this member condition modification information instruction member reduces and is also to increase, if
Being to reduce, proceed to step (4), then proceeding to step (7), without receiving then if increasing
Persistently monitor the member condition modification information from other member node；
(4) member node judges whether the heartbeat message received from father node or child node, if
Receive, encapsulate out a member condition modification information according to heartbeat message, and send it to center
Management node, then proceeds to step (5), and otherwise this member node continues to monitor from father node or son
The heartbeat message of node；
(5) central administration node judges that the member condition modification information from member node is the most correct,
If correct, update overall network topology structure, and the overall network topology structure after updating sends
To all of member node, then it is distributed to its descendants's member node by root member node, then turns
Enter step (6), otherwise wait for receiving the member condition modification information from other member node；
(6) network that all member node update its this locality according to the overall network topology figure after updating is opened up
Flutter structure；
(7) according to member condition modification information, member node determines whether that new member node request obtains
Take overall network topology structure, if then sending report information to central administration node, to ask centering
Management node is the member node distribution colouring information that this is new, then proceeds to step (8), otherwise member
Node continues to monitor whether there is member condition modification information；
(8) central administration node is this newcomer's node distribution colouring information according to report information, and more
The newly overall network topology structure of himself, and the overall network topology structure after new is sent to new
Member node, proceeds to step (9), and the overall network topology structure after the most also updating is distributed to institute
There is the root member node of tree, then have root member node to be distributed to its all of descendants's member node, proceed to
Step (6)；
(9) new member node is tied according to from the overall network topology after the renewal of central administration node
Structure updates the network topology structure of its this locality, and inquires about its father and son according to the network topology structure after updating
Member node information, and set up heart beating with its father and son's member node and contact.
Preferably, step (1) specifically includes following sub-step:
(1-1) central administration node in distributed type assemblies receives manager's joint all to distributed type assemblies
Name node1, the node2 of point ..., noden, wherein n is the sum of distributed type assemblies interior joint,
And by this name synchronizing information to all nodes；
(1-2) central administration node color set, maximal tree number and maximum child node to distributed type assemblies
Number configures, and as limiting parameter, configuration result is transmitted to all member node, for whole
The topological structure of individual distributed type assemblies is defined；
(1-3) color set is distributed to all nodes after distributed type assemblies starts by central administration node,
Then according to the maximal tree number and of configuration in color, IP address and the step (1-2) of each node
Big son node number builds the overall network topology figure being made up of many trees, as shown in Figure 1；
(1-4) central administration node is overall network topology configuration settings release format, initial version number
And incremental manner, and this overall network topology structure is distributed to the root member node of all trees, then
Root member node is had to be distributed to its descendants's member node；
Preferably, the node of numbered node1 is central administration node, and other node is that member saves
Point.
Preferably, in step (1-3), the building process of overall network topology follows following principle:
(1-3-1) node of same color concentrates on root node or the intermediate node of one tree, and dispersion
In other leaf node set；
(1-3-2) node can simultaneously appear in many trees, but each node can only be at one
Tree serves as root node or intermediate node；
(1-3-3) new node should be inserted into therewith with node near root node in the tree of color
In, and the number of this node child nodes is not less than maximum son node number；
(1-3-4) when the number of child nodes of a node reaches maximum son node number, newly inserted node
Should be inserted in the subtree of this node, and the subtree being inserted into be all subtree camber the shortest or
Person's descendant nodes minimum number；
(1-3-5) when in overall network topology structure, the sum of tree reaches maximal tree number, if had again
New node adds distributed type assemblies, does not creates new tree construction for this new node, but by this new joint
Point inserts in existing tree construction.
Preferably, in step (3), the heartbeat mechanism of certain member node of member node and other is unexpected
During interruption, show that this member node have received the member condition modification information that member reduces, member node
Have received the heartbeat message of non-existent new member node in the overall network topology structure that it is safeguarded
Time, show that this member node have received the member condition modification information that member increases.
Preferably, in step (5), become from more than one if central administration node is simultaneously received
The member condition modification information of member's node, then it represents that member condition modification information is correct, if only
Receive the member condition modification information from a member node, then it represents that member condition modification information
It is wrong.
In general, by the contemplated above technical scheme of the present invention compared with prior art, it is possible to
Obtain following beneficial effect:
1, favorable expandability: due to the fact that the member organization's mode and step that have employed in step (1)
(3) in, the AM automatic monitoring to member node change is machine-processed so that distributed coordination service system is having newly
Member add and old member can the capturing this change and make corresponding tune of Real-time and Dynamic when removing
Whole.It is thereby achieved that the dynamic expansion of distributed coordination service system.
2, reliability: owing to have employed the member organization's mode in step (1) and ordering with version number
The network topology structure of name, simplifies the process of the synchronizing information of whole coordination service system.This kind becomes
The crossing redundancy characteristic of member's organizational form interior joint so that the distribution of data is also redundancy, so working as
When having node failure, data also accurately and reliably can be synchronized to all member node.
3, high efficiency: in step (1) and step (5), the synchronization of all data and overall situation net
The distribution of network topological structure is all the root node being first distributed to each tree by central administration node, then by
Root node is distributed to its descendant nodes, shares with different levels for the load of central administration node to every
The root node of tree, it is to avoid central administration node is because loading excessive performance performance bottleneck.
4 optimalitys: in step (1), to right in overall network topology structure on central administration node
The quantity of tree and the maximum son node number of node limit, thus the height of the tree controlled and dimension,
Ensure that the load balancing of each tree, improve the overall performance of coordination service system.
Accompanying drawing explanation
Fig. 1 is the structural representation of the intermediate node of the present invention intersection woods.
Fig. 2 is the flow chart of extendible distributed coordination service management under cloud platform of the present invention.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with accompanying drawing
And embodiment, the present invention is further elaborated.Should be appreciated that described herein specifically
Embodiment only in order to explain the present invention, is not intended to limit the present invention.Additionally, it is disclosed below
Just may be used as long as technical characteristic involved in each embodiment of the present invention does not constitutes conflict each other
To be mutually combined.
The Integral Thought of the present invention is, the member of coordination service system is used intermediate node Cross-Tree
The organizational form of woods, simultaneously in the building process to coordination service system network topology, to tree
Height and dimension be defined, it is ensured that the load balancing that system is overall.Management process in system
In, father and son's internodal heart beating monitoring mechanism can make the perceiving of coordination service system Real-time and Dynamic to be
The change of system member node, thus realize the dynamic expansion of coordination service system.Data distribution and synchronization
The redundancy distribution of process ensure that system reliability.
Under cloud platform of the present invention, extendible distributed coordination service management comprises the following steps:
(1) at the beginning of all nodes of distributed type assemblies being carried out with the organizational form of the intermediate node intersection woods
Beginningization, specifically includes following sub-step:
(1-1) central administration node in distributed type assemblies receives manager's joint all to distributed type assemblies
Name node1, the node2 of point ..., noden, wherein n is the sum of distributed type assemblies interior joint,
And by this name synchronizing information to all nodes；Specifically, numbered node1 under default situations
Node is central administration node, and other node is member node；
(1-2) central administration node is to the color set (Color_Set) of distributed type assemblies, maximal tree number
(Max_Color_Num) configure with maximum son node number (Max_Children_Size), and
Configuration result is transmitted to all member node as limiting parameter, for whole distributed type assemblies
Topological structure is defined；Specifically, color set, maximal tree number and maximum son node number are by dividing
The scale of cloth cluster determines；
(1-3) color set is distributed to all nodes after distributed type assemblies starts by central administration node,
Then according to the maximal tree number and of configuration in color, IP address and the step (1-2) of each node
Big son node number builds the overall network topology figure being made up of many trees, as shown in Figure 1；Specifically,
The building process of overall network topology follows following principle:
(1-3-1) node of same color concentrates on root node or the intermediate node of one tree, and dispersion
In other leaf node set；
(1-3-2) node can simultaneously appear in many trees, but each node can only be at one
Tree serves as root node or intermediate node；
(1-3-3) new node should be inserted into therewith with node near root node in the tree of color
In, and the number of this node child nodes is not less than maximum son node number；
(1-3-4) when the number of child nodes of a node reaches maximum son node number, newly inserted node
Should be inserted in the subtree of this node, and the subtree being inserted into be all subtree camber the shortest or
Person's descendant nodes minimum number；
(1-3-5) when in overall network topology structure, the sum of tree reaches maximal tree number, if had again
New node adds distributed type assemblies, does not creates new tree construction for this new node, but by this new joint
Point inserts in existing tree construction；
(1-4) central administration node is overall network topology configuration settings release format, initial version number
And incremental manner, and this overall network topology structure is distributed to the root member node of all trees, then
Root member node is had to be distributed to its descendants's member node；
The advantage of step (1) is, it is to avoid the bottleneck problem that centralization organizational form is easily formed, with
Time limited by parameter and can ensure that the height of each tree and interstitial content approximation are consistent, also ensure that
Every tree load be the transaction time in approximate equalization, and each tree be similar, thus
Ensure that the efficiency that global data is submitted to.
(2) all member node go out according to the overall network topology structure query from central administration node
Father and son's member node information of this node, carries out information with heartbeat mechanism between all father and son's member node
Synchronize and communicate；
(3) member node judges whether to receive member condition modification information from other member node, as
Fruit receives, and determines whether that this member condition modification information instruction member reduces and is also to increase, if
Being to reduce, proceed to step (4), then proceeding to step (7), without receiving then if increasing
Persistently monitor the member condition modification information from other member node；Specifically, member node with
The heartbeat mechanism of other certain member node interrupts suddenly showing that this member node have received what member reduced
Member condition modification information, member node have received in the overall network topology structure that it is safeguarded and do not deposits
The heartbeat message of new member node time show that this member node have received member's shape that member increases
State modification information.
(4) member node judges whether that receiving the heartbeat message from father node or child node (i.e. exists
In topological structure, by heartbeat mechanism, its father node or child node report that it is raw within the time specified
Life existing state), if received, encapsulate out a member condition modification information according to heartbeat message,
And send it to central administration node, and then proceed to step (5), otherwise this member node continues prison
Listen the heartbeat message from father node or child node；
(5) central administration node judges that the member condition modification information from member node is the most correct,
If correct, update overall network topology structure, and the overall network topology structure after updating sends
To all of member node, then it is distributed to its descendants's member node by root member node, then turns
Enter step (6), otherwise wait for receiving the member condition modification information from other member node；Specifically
For, if the member condition that central administration node is simultaneously received from more than one member node becomes
More information, then it represents that member condition modification information is correct, if only received from a member
The member condition modification information of node, then it represents that member condition modification information is wrong；
The advantage of step (5) is, central administration node can be according to the member condition change received
The bar number of information judge this member condition modification information be owing to network condition is the best that cause or
Member node is really had to change, thus the correctness in the system expansion process ensured and reliability.
(6) network that all member node update its this locality according to the overall network topology figure after updating is opened up
Flutter structure；
(7) according to member condition modification information, member node determines whether that new member node request obtains
Take overall network topology structure, if then sending report information to central administration node, to ask centering
Management node is the member node distribution colouring information that this is new, then proceeds to step (8), otherwise member
Node continues to monitor whether there is member condition modification information；
(8) central administration node is this newcomer's node distribution colouring information according to report information, and more
The newly overall network topology structure of himself, and the overall network topology structure after new is sent to new
Member node, proceeds to step (9), and the overall network topology structure after the most also updating is distributed to institute
There is the root member node of tree, then have root member node to be distributed to its all of descendants's member node, proceed to
Step (6)；
The advantage of step (8) is, after overall network topology structure updates, for update
The distribution of overall network topology structure uses the distribution mechanisms of layering, can reduce central administration node
Load, reduces and postpones, and improves the efficiency of system.
(9) new member node is tied according to from the overall network topology after the renewal of central administration node
Structure updates the network topology structure of its this locality, and inquires about its father and son according to the network topology structure after updating
Member node information, and set up heart beating with its father and son's member node and contact.
As it will be easily appreciated by one skilled in the art that and the foregoing is only presently preferred embodiments of the present invention,
Not in order to limit the present invention, all made within the spirit and principles in the present invention any amendment, etc.
With replacement and improvement etc., should be included within the scope of the present invention.
Claims (6)
1. extendible distributed coordination service management under a cloud platform, it is characterised in that comprise the following steps:
(1) with the organizational form of the intermediate node intersection woods, all nodes of distributed type assemblies are initialized；
(2) all member node go out father and son's member node information of this node according to the overall network topology structure query from central administration node, carry out synchronizing information with heartbeat mechanism and communicate between all father and son's member node；
(3) member node judges whether to receive member condition modification information from other member node, if receiving, determining whether that this member condition modification information instruction member reduces and being also to increase, step (4) is then proceeded to if reducing, then proceeding to step (7) if increasing, without receiving, persistently monitoring the member condition modification information from other member node；
(4) member node judges whether the heartbeat message received from father node or child node, if received, encapsulate out a member condition modification information according to heartbeat message, and send it to central administration node, then proceeding to step (5), otherwise this member node continues to monitor from father node or the heartbeat message of child node；
(5) central administration node judges that the member condition modification information from member node is the most correct, if correct, update overall network topology structure, and the overall network topology structure after updating is sent to all of member node, its descendants's member node it is distributed to again by root member node, then proceed to step (6), otherwise wait for receiving the member condition modification information from other member node；
(6) all member node update the network topology structure of its this locality according to the overall network topology figure after updating；
(7) member node determines whether new member node acquisition request overall network topology structure according to member condition modification information, if then sending report information to central administration node, it is the member node distribution colouring information that this is new with request central administration node, then proceeding to step (8), otherwise member node continues to monitor whether there is member condition modification information；
(8) central administration node is that this newcomer's node distributes colouring information according to report information, and update the overall network topology structure of himself, and the overall network topology structure after updating is sent to new member node, proceed to step (9), the most also the overall network topology structure after updating is distributed to the root member node of all trees, it is distributed to its all of descendants's member node by root member node again, proceeds to step (6)；
(9) new member node is according to the network topology structure from its this locality of overall network topology topology update after the renewal of central administration node, and inquire about its father and son's member node information according to the network topology structure after updating, and set up heart beating with its father and son's member node and contact.
Distributed coordination service management the most according to claim 1, it is characterised in that step (1) specifically includes following sub-step:
(1-1) central administration node in distributed type assemblies receives name node1, the node2 of manager's node all to distributed type assemblies ..., noden, wherein n is the sum of distributed type assemblies interior joint, and by this name synchronizing information to all nodes；
(1-2) color set, maximal tree number and the maximum son node number of distributed type assemblies are configured by central administration node, and as limiting parameter, configuration result are transmitted to all member node, for being defined the topological structure of whole distributed type assemblies；
(1-3) color set is distributed to all nodes after distributed type assemblies starts by central administration node, then builds, according to maximal tree number and the maximum son node number of configuration in color, IP address and the step (1-2) of each node, the overall network topology figure being made up of many trees；
(1-4) central administration node is overall network topology configuration settings release format, initial version number and incremental manner, and this overall network topology structure is distributed to the root member node of all trees, then it is distributed to its descendants's member node by root member node.
Distributed coordination service management the most according to claim 1, it is characterised in that the node of numbered node1 is central administration node, other node is member node.
Distributed coordination service management the most according to claim 1, it is characterised in that in step (1-3), the building process of overall network topology follows following principle:
(1-3-1) node of same color concentrates on root node or the intermediate node of one tree, and is dispersed in the leaf node of other tree；
(1-3-2) node can simultaneously appear in many trees, but each node can only serve as root node or intermediate node in one tree；
(1-3-3) new node should be inserted into therewith with in the tree of color near in the node of root node, and the number of this node child nodes is not less than maximum son node number；
(1-3-4) when the number of child nodes of a node reaches maximum son node number, newly inserted node should be inserted in the subtree of this node, and the subtree being inserted into is that all subtree camber are the shortest or descendant nodes minimum number；
(1-3-5) when the sum of tree reaches maximal tree number in overall network topology structure, if there being new node to add distributed type assemblies again, not creating new tree construction for this new node, but this new node is inserted in existing tree construction.
Distributed coordination service management the most according to claim 1, it is characterized in that, in step (3), when the heartbeat mechanism of member node and other certain member node interrupts suddenly, show that this member node have received the member condition modification information that member reduces, when member node have received the heartbeat message of non-existent new member node in the overall network topology structure that it is safeguarded, show that this member node have received the member condition modification information that member increases.
Distributed coordination service management the most according to claim 1, it is characterized in that, in step (5), if central administration node is simultaneously received the member condition modification information from two or more member node, then represent that member's Status Change information is correct, if only receiving the member condition modification information from a member node, then it represents that member condition modification information is wrong.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310749438.5A CN103780497B (en) | 2013-12-30 | 2013-12-30 | Extendible distributed coordination service management under a kind of cloud platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310749438.5A CN103780497B (en) | 2013-12-30 | 2013-12-30 | Extendible distributed coordination service management under a kind of cloud platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103780497A CN103780497A (en) | 2014-05-07 |
| CN103780497B true CN103780497B (en) | 2016-08-10 |
Family
ID=50572350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310749438.5A Active CN103780497B (en) | 2013-12-30 | 2013-12-30 | Extendible distributed coordination service management under a kind of cloud platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103780497B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10069941B2 (en) | 2015-04-28 | 2018-09-04 | Microsoft Technology Licensing, Llc | Scalable event-based notifications |
| US10116767B2 (en) * | 2015-11-13 | 2018-10-30 | Furturewei Technologies, Inc. | Scaling cloud rendezvous points in a hierarchical and distributed manner |
| CN106302726B (en) * | 2016-08-15 | 2019-05-28 | 浪潮(北京)电子信息产业有限公司 | A kind of monitoring method and module of the management node of distributed type assemblies data |
| CN106444579B (en) * | 2016-10-09 | 2019-01-25 | 鼎点视讯科技有限公司 | A kind of frequency spectrum data multiplex system and method |
| CN107040466B (en) * | 2017-03-07 | 2020-09-08 | 西安电子科技大学 | Multi-domain cooperative data transmission path selection method based on layered architecture of Internet of things |
| CN107342942B (en) * | 2017-07-05 | 2020-09-18 | 苏州浪潮智能科技有限公司 | Information transmission method and device for multi-node controller cluster |
| CN110932875B (en) * | 2018-09-19 | 2022-11-22 | 中兴通讯股份有限公司 | Alarm device and method in big data cloud computing environment |
| US11075812B2 (en) * | 2019-06-20 | 2021-07-27 | Kaloom Inc. | Server and methods for synchronizing networking information with client devices |
| CN111865714B (en) * | 2020-06-24 | 2022-08-02 | 上海上实龙创智能科技股份有限公司 | Cluster management method based on multi-cloud environment |
| CN112565651B (en) * | 2020-11-30 | 2022-10-21 | 深圳数字电视国家工程实验室股份有限公司 | Data transmission method, electronic device and computer readable storage medium |
-
2013
- 2013-12-30 CN CN201310749438.5A patent/CN103780497B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103780497A (en) | 2014-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103780497B (en) | Extendible distributed coordination service management under a kind of cloud platform | |
| CN104008152B (en) | Support the framework method of the distributed file system of mass data access | |
| CN102739775B (en) | The monitoring of internet of things data acquisition server cluster and management method | |
| CN106170782B (en) | System and method for creating highly scalable high availability clusters in massively parallel processing clusters of machines in a network | |
| US20150067135A1 (en) | Member-oriented hybrid cloud operating system architecture and communication method thereof | |
| CN105703940A (en) | Multistage dispatching distributed parallel computing-oriented monitoring system and monitoring method | |
| CN108712464A (en) | A kind of implementation method towards cluster micro services High Availabitity | |
| CN104301391B (en) | Multi-area optical network data center resource virtualizes mapping method | |
| CN107579931A (en) | A kind of block chain based on Kubernetes is Service Source adaptation method | |
| CN103581276B (en) | Cluster management device, system, service customer end and correlation method | |
| CN105959349B (en) | A kind of Distributed Services end operating system and method | |
| CN104320459A (en) | Node management method and device | |
| US9571299B2 (en) | Method for managing tree structure, information processing system, and medium | |
| CN103581332B (en) | HDFS framework and pressure decomposition method for NameNodes in HDFS framework | |
| CN102984501A (en) | Network video-recording cluster system | |
| CN107124306A (en) | Content delivery network server optimization dispositions method under network function virtualized environment | |
| CN106302569A (en) | Process method and the computer system of cluster virtual machine | |
| CN106301822A (en) | A kind of methods, devices and systems that AP is configured | |
| CN106713391A (en) | Session information sharing method and sharing system | |
| CN109639773A (en) | A kind of the distributed data cluster control system and its method of dynamic construction | |
| CN108512865B (en) | Data transmission method and device of EJB container system of Java application server | |
| CN110334072A (en) | A kind of distributed file system, file updating method and device | |
| CN107682411A (en) | A kind of extensive SDN controllers cluster and network system | |
| CN108228393A (en) | A kind of implementation method of expansible big data High Availabitity | |
| CN110213359A (en) | A kind of car networking networking data delivery system and method based on D2D |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |