CN105812425A - Method for establishing cross-room server cluster and node servers - Google Patents
Method for establishing cross-room server cluster and node servers Download PDFInfo
- Publication number
- CN105812425A CN105812425A CN201410850872.7A CN201410850872A CN105812425A CN 105812425 A CN105812425 A CN 105812425A CN 201410850872 A CN201410850872 A CN 201410850872A CN 105812425 A CN105812425 A CN 105812425A
- Authority
- CN
- China
- Prior art keywords
- server
- nodal point
- node
- primary nodal
- point server
- 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.)
- Granted
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention belongs to the communication field and provides a method for establishing a cross-room server cluster and node servers. The method comprises steps that the member information is configured by a first node server; Ping packets are respectively sent to members through a transmission level TCP protocol; return packets which include member views and are sent by a second node server are received through the transmission level TCP protocol; the member views are acquired from the return packets; whether a present node server is a group leader is determined according to the member views acquired from the return packets sent by all the members, if not, a message for requesting the group leader to join in the server cluster is sent to the group leader; if yes, other node servers are waited to join, and member views are updated; a member view update event is sent by the first node server to all the members in the server cluster. According to the method, the node servers for cloud service are expanded to the WAN environment, access of the node servers in different areas is realized, cloud service nodes are dynamically expanded, and simultaneous online capacity of the servers is improved.
Description
Technical field
The invention belongs to the communications field, particularly relate to a kind of establishment across the method for server in machine room cluster and node server.
Background technology
Nearly 2 years, new technique (the Internet, mobile Internet, Wearable etc.) was not content with the field of oneself, starts front on a large scale and marches traditional industries.The Internet essentially, is the means by technology, the problem solving information asymmetry.What it substantially solved is the problem of " communication ", and it is originally wanting cross-region, all evening up across time, the state that cannot link up across language.Therefore the Internet also greatly releases efficiency.The change that the Internet brings is in that, links up the very fast lifting of efficiency.This lifting, by the mode eliminating intermediate link, reconstructs commercial value chain.
Industry is not regarded as the Internet individually can change the reconstruct of whole industry, but have four bursts of strength, these four bursts of strength include movement, social, cloud, big data, these four bursts of strength are occured simultaneously together, change the behavior of consumer, defining the ecosystem that consumer promotes, this ecosystem changes enterprise further and does business and the mode of industry reconstruct.
Current IT industry mainly seizes highly competitive at Internet portal.Internet portal is just above the product of user's request, as the door type website in PC epoch, the blog in Web2.0 epoch, the wechat in mobile Internet epoch till now.
Support a product exceeding user's request, it is necessary to powerful high in the clouds supports.Build a powerful high in the clouds, it is necessary to build a stable server cluster.Server cluster refers to get up a lot of server centered carry out same service together, similarly is only one of which server in client.Cluster can utilize multiple computer to carry out parallel computation thus obtaining significantly high calculating speed, it is also possible to backups with multiple computers so that any one machine be broken whole system still can be properly functioning.
Build server cluster, it is necessary to be fully understood by TCP/IP architecture and agreement, configure a suitable protocol stack.Such as, host-host protocol is the agreement of the protocol stack bottom, and it is responsible for receiving message from physical network environment, the message received is delivered to the agreement that protocol stack is adjacent at the middle and upper levels, and the message that protocol stack agreement at the middle and upper levels is transmitted is sent to physical network.Transport layer protocol mainly includes UDP, TCP, TUNNEL.The node server that UDP is generally used in cluster is distributed in the situation in LAN, uses IPmulticast to send message to cluster, uses UDPdatagram to send unicast message to single node server.When the node server in cluster is distributed in WAN, router can abandon IPmulticast message, and therefore UDP is not suitable for setting up across server in machine room cluster.Therefore prior art cannot realize the node server of cloud service is extended to WAN environment, it is achieved the node server of different regions accesses.
Summary of the invention
It is an object of the invention to provide a kind of establishment across the method for server in machine room cluster and node server, it is intended to solve prior art and cannot realize the node server of cloud service is extended to WAN environment, it is achieved the problem that the node server of different regions accesses.
First aspect, the invention provides a kind of establishment method across server in machine room cluster, and described method includes:
Primary nodal point server configures information about firms, described member includes other node server of this machine room and the interface server of other machine room；
Primary nodal point server sends Ping bag to described member respectively by transport layer Transmission Control Protocol；
Primary nodal point server receives, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends；
Primary nodal point server obtains member view wrapping from described time；
The member view obtained in bag that returns that primary nodal point server sends according to all members judges whether this node server is group leader, if it is not, then primary nodal point server sends the message adding server cluster to group leader；If it is, primary nodal point server waits that other node servers add, and update member view；
Primary nodal point server sends member view's more new events to all members in server cluster.
Second aspect, the invention provides a kind of establishment method across server in machine room cluster, and described method includes:
Secondary nodal point server receives Ping bag by transport layer Transmission Control Protocol, described Ping bag is sent to its member by transport layer Transmission Control Protocol respectively by primary nodal point server, described member is by primary nodal point server configures, and described member includes other node server of the machine room at primary nodal point server place and the interface server of other machine room；
Secondary nodal point server obtains the physical address of this node server；
Physical address and the member view of this node server are encapsulated in back in bag by secondary nodal point server；
Secondary nodal point server sends time bag comprising member view by transport layer Transmission Control Protocol to primary nodal point server, so that primary nodal point server adds server cluster, or update member view, and send member view's more new events to all members in server cluster.
The third aspect, the invention provides a kind of node server, and described node server includes:
Configuration module, for configuration member's information, described member includes other node server of this machine room and the interface server of other machine room；
First sending module, for sending Ping bag to described member respectively by transport layer Transmission Control Protocol；
First receiver module, for receiving, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends；
First acquisition module, for obtaining member view from described time wrapping；
For the member view obtained in bag that returns sent according to all members, judge module, judges whether this node server is group leader；
Second sending module, for when described judge module judges that this node server is not group leader, sending the message adding server cluster to group leader；
More new module, for when described judge module judges that this node server is group leader, waiting that other node servers add, and update member view；
3rd sending module, for sending member view's more new events to all members in server cluster.
Fourth aspect, the invention provides a kind of node server, and described node server includes:
Second receiver module, for receiving Ping bag by transport layer Transmission Control Protocol, described Ping bag is sent to its member by transport layer Transmission Control Protocol respectively by primary nodal point server, described member is by primary nodal point server configures, and described member includes other node server of the machine room at primary nodal point server place and the interface server of other machine room；
Second acquisition module, for obtaining the physical address of this node server；
Second package module, for being encapsulated in back the physical address of this node server and member view in bag；
4th sending module, for being sent time bag comprising member view to primary nodal point server by transport layer Transmission Control Protocol, so that primary nodal point server adds server cluster, or update member view, and send member view's more new events to all members in server cluster.
In the present invention, owing to adopting transport layer Transmission Control Protocol to be transmitted between node server, it is achieved that the node server of cloud service is extended to WAN environment, it is achieved that the node server of different regions accesses, dynamic expansion cloud service node, improves server online capacity simultaneously.
Accompanying drawing explanation
Fig. 1 is server cluster architectural schematic.
Fig. 2 is the physical structure figure across server in machine room cluster to be set up in the embodiment of the present invention.
Fig. 3 is the establishment method flow diagram across server in machine room cluster of the embodiment of the present invention one offer.
Fig. 4 is the node server structural representation that the embodiment of the present invention two provides.
Fig. 5 is the node server structural representation that the embodiment of the present invention three provides.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and beneficial effect clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein is only in order to explain the present invention, is not intended to limit the present invention.
In order to technical solutions according to the invention are described, illustrate below by specific embodiment.
Refer to Fig. 1, be the server cluster architectural schematic of three node servers.In protocol stack on transport layer protocol (such as TCP) physically based deformation network, passage (Channel) is based on protocol stack, and application (Application) is based on passage.
In server cluster, member can carry out communication point-to-point, that point is to group, and all communications are completed by passage.When each member creates passage, a protocol stack (ProtocolStack) to be configured to passage.Passage is based on protocol stack.
Protocol stack includes following agreement:
TCP: send message by multicast (Multicast) agreement based on UDP
TCPPING: send message segment (Segment) to other member by transport layer Transmission Control Protocol, attempts the logical the other side of ping
MERGE2: merge subgroup, or a sub-component from time remove subgroup
FD: based on the error detection protocol of heart beating (Heartbeat) message, can generate notice when certain member's failure
Pbcast.NAKACK: guaranteeing the reliable of message and FIFO order, message reliability ensure that a message can be received, and otherwise recipient may require that re-transmission, FIFO ensure that all message sent from sender can arrive recipient with same sequence
UNICAST: clean culture
Pbcast.STABLE: the message received by all members is deleted
Pbcast.GMS: Membership Protocol, the addition of responsible member/leave and update new member view
FRAG2: huge message is smashed the transmission being divided into several little, and re-assemblies back at receiving terminal, it is adaptable to multicast and unicast protocol
Pbcast.STATE_TRANSFER: state transfer, it is ensured that state correctly can be transferred to other member from a member
Forming doubly linked list in protocol stack between each layer protocol, namely every layer protocol has upper-layer protocol (up_protocol) and lower-layer protocols (down_protocol), it is possible to bi-directional message.
Referring to Fig. 2, the embodiment of the present invention is had 4 different machine rooms with to set up across server in machine room cluster, i.e. 4 different Internet data centers (InternetDataCenter, IDC): IDC1, IDC2, IDC3 and IDC4, for example.Wherein, IDC1 has multiple stage node server, such as 121.201.96.190,121.201.96.191,121.201.96.192 etc.；IDC2 has multiple stage node server, such as 218.17.160.100,218.17.160.101,218.17.160.102 etc.；IDC3 has multiple stage node server, such as 61.174.9.250,61.174.9.251,61.174.9.252 etc.；IDC4 has multiple stage node server, such as 120.31.134.50,120.31.134.51,120.31.134.52 etc..Each machine room specifies a node server as interface server, and interface server is configured with the information of other node server of this machine room, simultaneously that all node servers of other machine room are visible.
The method across server in machine room cluster of setting up that these 12 node servers are provided by the embodiment of the present invention one shown in Fig. 3 is set up.
Embodiment one:
Referring to Fig. 3, the establishment that the embodiment of the present invention one provides comprises the following steps across the method for server in machine room cluster:
S101, primary nodal point server configures information about firms, described member includes other node server of this machine room and the interface server of other machine room；
In the embodiment of the present invention one, if primary nodal point server is interface server, then other node server of this machine room refers to the node server of other non-interface server；If primary nodal point server is not interface server, then other node server of this machine room includes the interface server of this machine room and the node server of other non-interface server.
Such as: in conjunction with Fig. 2,
1) configuration of interface server 1:
a)bind_port:7800
B) initial_hosts:
121.201.96.192:7800, (this node server ip:port)
218.17.160.100:7800,61.174.9.252:7800,120.31.134.50:7800, (the interface server ip:port of other machine room)
(121.201.96.190:7800,121.201.96.191:7800 the ip:port of other node server of IDC1)
2) configuration of the node server of non-interface server: select as an example, as 121.201.96.190 this.
a)bind_port:7800
B) initial_hosts:
121.201.96.190:7800, (this node server ip:port)
121.201.96.192:7800,218.17.160.100:7800,61.174.9.252:780 0,120.31.134.50:7800 (ip:port of the interface server of this machine room and the interface server of other machine room)
121.201.96.191:7800, (ip:port of other node server of IDC1)
3) configuration of interface server 2:
a)bind_port:7800
B) initial_hosts:
218.17.160.100:7800,
121.201.96.192:7800,61.174.9.252:7800,120.31.134.50:7800,
218.17.160.101:7800,218.17.160.102:7800
4) configuration of interface server 3:
a)bind_port:7800
B) initial_hosts:
61.174.9.252:7800,
121.201.96.192:7800,218.17.160.100:7800,120.31.134.50:7800,
61.174.9.250:7800,61.174.9.251:7800
5) configuration of interface server 4:
a)bind_port:7800
B) initial_hosts:
120.31.134.50:7800,
121.201.96.192:7800,218.17.160.100:7800,61.174.9.252:7800,
120.31.134.51:7800,120.31.134.52:7800
S102, primary nodal point server send Ping bag to described member respectively by transport layer Transmission Control Protocol；
S103, secondary nodal point server receive described Ping bag by transport layer Transmission Control Protocol；
S104, secondary nodal point server obtain the physical address of this node server；
Physical address and the member view of this node server are encapsulated in back in bag by S105, secondary nodal point server；
S106, secondary nodal point server send time bag comprising member view by transport layer Transmission Control Protocol to primary nodal point server；
In the embodiment of the present invention one, return bag and specifically include following information attribute:
A) Addresssender: the place node address of sender
B) view: view object
Id: view number
Address [] members: members list
Addresscreator: view founder
C) Stringlogical_name: the place node logical title of sender
D) Collection<PhysicalAddress>physical_addrs: the address (containing physical address) of other member that can collect when sender adds cluster
S107, primary nodal point server receive, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends；
S108, primary nodal point server obtain member view wrapping from described time；
The member view obtained in bag that returns that S109, primary nodal point server send according to all members judges whether this node server is group leader, if it is not, then perform S110, if it is, perform S111；
S110, primary nodal point server send the message adding server cluster to group leader；
S111, primary nodal point server wait that other node servers add, and update member view；
S112, primary nodal point server send member view's more new events to all members in server cluster.
In the embodiment of the present invention one, S101 specifically may comprise steps of:
Primary nodal point server initiates connection event at channel layer, give group membership agreement (GroupMembershipProtocol, GMS) layer processes, GMS protocol specification: Handlesjoins/leaves/crashes (suspicions) andemitsnewviewsaccordingly. process adds/leaves/collapses (suspection), and send corresponding new view；
Primary nodal point server processes described connection event at GMS layer, sends lookup member's event to discovery layer (Discovery layer)；
Primary nodal point server processes in discovery layer and searches member's event；
Primary nodal point server is in discovery layer configuration member's information, and described member includes other node server of this machine room and the interface server of other machine room.
In the embodiment of the present invention one, S102 specifically may comprise steps of:
Primary nodal point server initiates request event to described member by transport layer Transmission Control Protocol in discovery layer；
Request event is encapsulated in universal information (Event.MSG) in discovery layer and is passed down to transport layer by primary nodal point server；
Primary nodal point server sends request event to described member in transport layer respectively by transport layer Transmission Control Protocol.
In the embodiment of the present invention one, S103 is particularly as follows: secondary nodal point server receives described request event in transport layer by transport layer Transmission Control Protocol.
In the embodiment of the present invention one, S104 specifically may comprise steps of:
Secondary nodal point server processes request event in discovery layer；
Secondary nodal point server obtains the physical address of this node server in discovery layer.
In the embodiment of the present invention one, S105 is passed down to transport layer particularly as follows: secondary nodal point server is encapsulated in back in bag at the discovery layer physical address by this node server and member view.
In the embodiment of the present invention one, S106 is particularly as follows: secondary nodal point server sends time bag comprising member view in transport layer by transport layer Transmission Control Protocol to primary nodal point server.
In the embodiment of the present invention one, S107 is particularly as follows: primary nodal point server receives, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends in transport layer.
In the embodiment of the present invention one, S108 particularly as follows: primary nodal point server discovery layer receive transport layer be communicated up described in return bag, and be passed up to GMS layer, until GMS layer receive that all members send return bag.
In the embodiment of the present invention one, S109 specifically includes following steps:
The member view obtained in bag that returns that primary nodal point server sends according to all members at GMS layer elects group leader；Particularly as follows: the member view obtained in bag that returns that primary nodal point server sends according to all members at GMS layer adds up member view founder, founder's election that the member view of establishment is maximum is group leader；
Judge whether this node server is group leader, if it is not, then perform S110, if it is, perform S111.
In the embodiment of the present invention one, described renewal member view specifically includes following steps:
Group leader integrates the view members list returned in bag that all members send, and generates the up-to-date members list of cluster；If the member of configuration does not contain few members, then few members is added in dynamic member list, generate the up-to-date members list of cluster of the member including dynamic member list and configuration.
Group leader produces a new view, comprises the up-to-date members list of described cluster.
In the embodiment of the present invention one, owing to adopting transport layer Transmission Control Protocol to be transmitted between node server, even if the member therefore in server cluster is distributed in WAN, server cluster also can be set up, dynamic expansion cloud service node, improves server online capacity simultaneously.
Embodiment two:
Referring to Fig. 4, the node server that the embodiment of the present invention two provides includes: configuration module the 11, first sending module the 12, first receiver module the 13, first acquisition module 14, judge module the 15, second sending module 16, more new module 17 and the 3rd sending module 18.
Configuration module 11, for configuration member's information, described member includes other node server of this machine room and the interface server of other machine room；
In the embodiment of the present invention two, if primary nodal point server is interface server, then other node server of this machine room refers to the node server of other non-interface server；If primary nodal point server is not interface server, then other node server of this machine room includes the interface server of this machine room and the node server of other non-interface server.
First sending module 12, for sending Ping bag to described member respectively by transport layer Transmission Control Protocol；
First receiver module 13, for receiving, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends；
First acquisition module 14, for obtaining member view from described time wrapping；
For the member view obtained in bag that returns sent according to all members, judge module 15, judges whether this node server is group leader；
Second sending module 16, for when described judge module judges that this node server is not group leader, sending the message adding server cluster to group leader；
More new module 17, for when described judge module judges that this node server is group leader, waiting that other node servers add, and update member view；
3rd sending module 18, for sending member view's more new events to all members in server cluster.
In the embodiment of the present invention two, described configuration module 11 specifically includes:
First initiation module, for initiating connection event at channel layer, gives GMS layer and processes；
First processing module, for processing described connection event at GMS layer, sends lookup member's event to discovery layer；
Second processing module, searches member's event for processing in discovery layer；
Configuration submodule, in discovery layer configuration member's information, described member includes other node server of this machine room and the interface server of other machine room；
Described first sending module 12 specifically includes:
Second initiation module, for initiating request event to described member by transport layer Transmission Control Protocol in discovery layer；
First package module, for being encapsulated in universal information in discovery layer by request event and being passed down to transport layer；
First sends submodule, for sending request event to described member in transport layer respectively by transport layer Transmission Control Protocol.
In the embodiment of the present invention two, described judge module 15 specifically includes:
Election module, elects group leader for the member view obtained in bag that returns sent according to all members at GMS layer；Particularly as follows: the member view obtained in time bag that GMS layer sends according to all members adds up member view founder, founder's election that the member view of establishment is maximum is group leader；
Judge submodule, be used for judging whether this node server is group leader.
In the embodiment of the present invention two, the first receiver module 13 is specifically for receiving, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends in transport layer.
In the embodiment of the present invention two, the first acquisition module 14 returns bag described in being communicated up in discovery layer reception transport layer, and is passed up to GMS layer, until GMS layer receives returning of all members transmission and wraps.
In the embodiment of the present invention two, described more new module 17 integrates, specifically for group leader, the view members list returned in bag that all members send, and generates the up-to-date members list of cluster；Group leader produces a new view, comprises the up-to-date members list of described cluster.
In the embodiment of the present invention two, owing to adopting transport layer Transmission Control Protocol to be transmitted between node server, even if the member therefore in server cluster is distributed in WAN, server cluster also can be set up, dynamic expansion cloud service node, improves server online capacity simultaneously.
Embodiment three:
Referring to Fig. 5, the node server that the embodiment of the present invention three provides includes: the second receiver module the 21, second acquisition module the 22, second package module 23 and the 4th sending module 24, wherein,
Second receiver module 21, for receiving Ping bag by transport layer Transmission Control Protocol, described Ping bag is sent to its member by transport layer Transmission Control Protocol respectively by primary nodal point server, described member is by primary nodal point server configures, and described member includes other node server of the machine room at primary nodal point server place and the interface server of other machine room；
Second acquisition module 22, for obtaining the physical address of this node server；
Second package module 23, for being encapsulated in back the physical address of this node server and member view in bag；
4th sending module 24, for being sent time bag comprising member view to primary nodal point server by transport layer Transmission Control Protocol, so that primary nodal point server adds server cluster, or update member view, and send member view's more new events to all members in server cluster.
In the embodiment of the present invention three, return bag and specifically include following information attribute:
A) Addresssender: the place node address of sender
B) view: view object
Id: view number
Address [] members: members list
Addresscreator: view founder
C) Stringlogical_name: the place node logical title of sender
D) Collection<PhysicalAddress>physical_addrs: the address (containing physical address) of other member that can collect when sender adds cluster
In the embodiment of the present invention three, the second receiver module 21 is specifically for receiving request event in transport layer by transport layer Transmission Control Protocol, and described request event is sent to its member by transport layer Transmission Control Protocol in transport layer respectively by primary nodal point server.
In the embodiment of the present invention three, the second acquisition module 22, specifically for processing request event in discovery layer, obtains the physical address of this node server in discovery layer.
In the embodiment of the present invention three, the second package module 23 is passed down to transport layer specifically for being encapsulated in back in bag at the discovery layer physical address by this node server and member view.
In the embodiment of the present invention three, the 4th sending module 24 specifically for sending time bag comprising member view in transport layer by transport layer Transmission Control Protocol to primary nodal point server.
In the embodiment of the present invention three, owing to adopting transport layer Transmission Control Protocol to be transmitted between node server, even if the member therefore in server cluster is distributed in WAN, server cluster also can be set up, dynamic expansion cloud service node, improves server online capacity simultaneously.
One of ordinary skill in the art will appreciate that all or part of step realizing in above-described embodiment method can be by the hardware that program carrys out instruction relevant and completes, described program can be stored in a computer read/write memory medium, described storage medium, such as ROM/RAM, disk, CD etc..
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all any amendment, equivalent replacement and improvement etc. made within the spirit and principles in the present invention, should be included within protection scope of the present invention.
Claims (10)
1. set up the method across server in machine room cluster for one kind, it is characterised in that described method includes:
Primary nodal point server configures information about firms, described member includes other node server of this machine room and the interface server of other machine room；
Primary nodal point server sends Ping bag to described member respectively by transport layer Transmission Control Protocol；
Primary nodal point server receives, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends；
Primary nodal point server obtains member view wrapping from described time；
The member view obtained in bag that returns that primary nodal point server sends according to all members judges whether this node server is group leader, if it is not, then primary nodal point server sends the message adding server cluster to group leader；If it is, primary nodal point server waits that other node servers add, and update member view；
Primary nodal point server sends member view's more new events to all members in server cluster.
2. the method for claim 1, it is characterised in that described primary nodal point server configures information about firms specifically includes:
Primary nodal point server initiates connection event at channel layer, gives GMS layer and processes；
Primary nodal point server processes described connection event at GMS layer, sends lookup member's event to discovery layer；
Primary nodal point server processes in discovery layer and searches member's event；
Primary nodal point server is in discovery layer configuration member's information, and described member includes other node server of this machine room and the interface server of other machine room.
3. the method for claim 1, it is characterised in that described primary nodal point server sends Ping bag to described member respectively by transport layer Transmission Control Protocol and specifically includes:
Primary nodal point server initiates request event to described member by transport layer Transmission Control Protocol in discovery layer；
Request event is encapsulated in universal information in discovery layer and is passed down to transport layer by primary nodal point server；
Primary nodal point server sends request event to described member in transport layer respectively by transport layer Transmission Control Protocol.
4. the method for claim 1, it is characterised in that the member view obtained in bag that returns that described primary nodal point server sends according to all members judges whether this node server is that group leader specifically includes:
The member view obtained in bag that returns that primary nodal point server sends according to all members at GMS layer elects group leader；Particularly as follows: the member view obtained in bag that returns that primary nodal point server sends according to all members at GMS layer adds up member view founder, founder's election that the member view of establishment is maximum is group leader；
Judge whether this node server is group leader.
5. set up the method across server in machine room cluster for one kind, it is characterised in that described method includes:
Secondary nodal point server receives Ping bag by transport layer Transmission Control Protocol, described Ping bag is sent to its member by transport layer Transmission Control Protocol respectively by primary nodal point server, described member is by primary nodal point server configures, and described member includes other node server of the machine room at primary nodal point server place and the interface server of other machine room；
Secondary nodal point server obtains the physical address of this node server；
Physical address and the member view of this node server are encapsulated in back in bag by secondary nodal point server；
Secondary nodal point server sends time bag comprising member view by transport layer Transmission Control Protocol to primary nodal point server, so that primary nodal point server adds server cluster, or update member view, and send member view's more new events to all members in server cluster.
6. method as claimed in claim 5, it is characterised in that described time bag specifically includes following information attribute:
The address of other member that the place node address of sender, view object, the place node logical title of sender and sender can collect when adding cluster, described view object includes view number, members list and view founder.
7. a node server, it is characterised in that described node server includes:
Configuration module, for configuration member's information, described member includes other node server of this machine room and the interface server of other machine room；
First sending module, for sending Ping bag to described member respectively by transport layer Transmission Control Protocol；
First receiver module, for receiving, by transport layer Transmission Control Protocol, time bag comprising member view that secondary nodal point server sends；
First acquisition module, for obtaining member view from described time wrapping；
For the member view obtained in bag that returns sent according to all members, judge module, judges whether this node server is group leader；
Second sending module, for when described judge module judges that this node server is not group leader, sending the message adding server cluster to group leader；
More new module, for when described judge module judges that this node server is group leader, waiting that other node servers add, and update member view；
3rd sending module, for sending member view's more new events to all members in server cluster.
8. node server as claimed in claim 7, it is characterised in that
Described configuration module specifically includes:
First initiation module, for initiating connection event at channel layer, gives GMS layer and processes；
First processing module, for processing described connection event at GMS layer, sends lookup member's event to discovery layer；
Second processing module, searches member's event for processing in discovery layer；
Configuration submodule, in discovery layer configuration member's information, described member includes other node server of this machine room and the interface server of other machine room；
Described first sending module specifically includes:
Second initiation module, for initiating request event to described member by transport layer Transmission Control Protocol in discovery layer；
First package module, for being encapsulated in universal information in discovery layer by request event and being passed down to transport layer；
First sends submodule, for sending request event to described member in transport layer respectively by transport layer Transmission Control Protocol.
9. node server as claimed in claim 7, it is characterised in that described judge module specifically includes:
Election module, elects group leader for the member view obtained in bag that returns sent according to all members at GMS layer；Particularly as follows: the member view obtained in time bag that GMS layer sends according to all members adds up member view founder, founder's election that the member view of establishment is maximum is group leader；
Judge submodule, be used for judging whether this node server is group leader.
10. a node server, it is characterised in that described node server includes:
Second receiver module, for receiving Ping bag by transport layer Transmission Control Protocol, described Ping bag is sent to its member by transport layer Transmission Control Protocol respectively by primary nodal point server, described member is by primary nodal point server configures, and described member includes other node server of the machine room at primary nodal point server place and the interface server of other machine room；
Second acquisition module, for obtaining the physical address of this node server；
Second package module, for being encapsulated in back the physical address of this node server and member view in bag；
4th sending module, for being sent time bag comprising member view to primary nodal point server by transport layer Transmission Control Protocol, so that primary nodal point server adds server cluster, or update member view, and send member view's more new events to all members in server cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410850872.7A CN105812425B (en) | 2014-12-30 | 2014-12-30 | A kind of method and node server of the establishment across server in machine room cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410850872.7A CN105812425B (en) | 2014-12-30 | 2014-12-30 | A kind of method and node server of the establishment across server in machine room cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105812425A true CN105812425A (en) | 2016-07-27 |
| CN105812425B CN105812425B (en) | 2019-04-19 |
Family
ID=56421566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410850872.7A Active CN105812425B (en) | 2014-12-30 | 2014-12-30 | A kind of method and node server of the establishment across server in machine room cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105812425B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1410905A (en) * | 2002-11-14 | 2003-04-16 | 华中科技大学 | Full distribution type aggregation network servicer system |
| CN101841541A (en) * | 2010-04-30 | 2010-09-22 | 浪潮(北京)电子信息产业有限公司 | Method and system for monitoring cluster based on multicast network |
| CN102624911A (en) * | 2012-03-14 | 2012-08-01 | 中山大学 | Cluster-based visible media storage system |
| CN103856554A (en) * | 2012-12-04 | 2014-06-11 | 国际商业机器公司 | Clustering support across geographical boundaries |
-
2014
- 2014-12-30 CN CN201410850872.7A patent/CN105812425B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1410905A (en) * | 2002-11-14 | 2003-04-16 | 华中科技大学 | Full distribution type aggregation network servicer system |
| CN101841541A (en) * | 2010-04-30 | 2010-09-22 | 浪潮(北京)电子信息产业有限公司 | Method and system for monitoring cluster based on multicast network |
| CN102624911A (en) * | 2012-03-14 | 2012-08-01 | 中山大学 | Cluster-based visible media storage system |
| CN103856554A (en) * | 2012-12-04 | 2014-06-11 | 国际商业机器公司 | Clustering support across geographical boundaries |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105812425B (en) | 2019-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Glaroudis et al. | Survey, comparison and research challenges of IoT application protocols for smart farming | |
| EP2705645B1 (en) | Name-based neighbor discovery and multi-hop service discovery in information-centric networks | |
| US9197721B2 (en) | Learning a MAC address | |
| US9705799B2 (en) | Server-side load balancing using parent-child link aggregation groups | |
| US7978631B1 (en) | Method and apparatus for encoding and mapping of virtual addresses for clusters | |
| CN103125102B (en) | For providing the system and method for the Ethernet virtual concentrator scalability based on infinite bandwidth in middleware machine environment | |
| US9413652B2 (en) | Systems and methods for path maximum transmission unit discovery | |
| EP1742430A1 (en) | Router redundancy in data communication networks | |
| KR101604970B1 (en) | Finding services in a service-oriented architecture(soa) network | |
| JP2017537547A (en) | Stateful load balancing in stateless networks | |
| CN105981347A (en) | System and method for preventing denial of service (DOS) attack on and for supporting reliable connection (RC) based on subnet administrator (SA) access in an engineered system for middleware and application execution | |
| Antonini et al. | Lightweight multicast forwarding for service discovery in low-power IoT networks | |
| WO2022253087A1 (en) | Data transmission method, node, network manager, and system | |
| CN107872368B (en) | Method and device for detecting accessibility of gateway in network node cluster and terminal | |
| Jones et al. | Protocol design for large group multicasting: the message distribution protocol | |
| US8345576B2 (en) | Methods and systems for dynamic subring definition within a multi-ring | |
| JP2022529459A (en) | Efficient message transmission and loop avoidance in RPL networks | |
| Altaweel et al. | Rsock: A resilient routing protocol for mobile fog/edge networks | |
| CN103107942B (en) | The tracking of a kind of static routing and equipment | |
| WO2013116384A1 (en) | Hierarchical network with active redundant links | |
| CN110995609A (en) | Message sending method and device, electronic equipment and storage medium | |
| CN105812425A (en) | Method for establishing cross-room server cluster and node servers | |
| Chae et al. | Fast discovery scheme using DHT-like overlay network for a large-scale DDS | |
| Antonielli | Development and comparison of MQTT distributed algorithms for HiveMQ | |
| US9398487B2 (en) | System and method for management of network links by traffic type |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |