CN110502342B - Maneuvering edge information service network in intermittent network environment - Google Patents

Maneuvering edge information service network in intermittent network environment Download PDF

Info

Publication number
CN110502342B
CN110502342B CN201910758332.9A CN201910758332A CN110502342B CN 110502342 B CN110502342 B CN 110502342B CN 201910758332 A CN201910758332 A CN 201910758332A CN 110502342 B CN110502342 B CN 110502342B
Authority
CN
China
Prior art keywords
task
node
network
manager
nodes
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
Application number
CN201910758332.9A
Other languages
Chinese (zh)
Other versions
CN110502342A (en
Inventor
李新明
刘斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Edge Intelligence Of Cas Co ltd
Original Assignee
Edge Intelligence Of Cas Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Edge Intelligence Of Cas Co ltd filed Critical Edge Intelligence Of Cas Co ltd
Priority to CN201910758332.9A priority Critical patent/CN110502342B/en
Publication of CN110502342A publication Critical patent/CN110502342A/en
Application granted granted Critical
Publication of CN110502342B publication Critical patent/CN110502342B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
    • G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/161—Computing infrastructure, e.g. computer clusters, blade chassis or hardware partitioning
    • 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
    • H04L41/04—Network management architectures or arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/70—Services for machine-to-machine communication [M2M] or machine type communication [MTC]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Abstract

The invention discloses a mobile edge information service network, which comprises a plurality of nodes, wherein the nodes are connected with each other in a wireless communication mode, and the mobile edge information service network comprises the following steps: the nodes comprise general nodes and main nodes, each node is provided with a task scheduling management system, the task scheduling management system comprises a task manager, a task executor, a task information collector and a task scheduler, the task manager and the task executor can operate on each node, and the task information collector and the task scheduler only operate on the main node.

Description

Maneuvering edge information service network in intermittent network environment
Technical Field
The present invention relates to computer networks, and more particularly, to a mobile edge information service network and control method in an intermittent network environment.
Background
Cloud computing technology is mature at a remarkable speed, and basic services such as simple computing and storage are expanded to various enterprise-level application service levels in a service mode. Under certain conditions, the computing ecological environment is complex and changeable, for example, under a mobile edge network, the mobile edge network is mostly formed by a plurality of nodes in discontinuous movement or continuous movement in a wireless connection mode, and often operates in occasions with severe communication environments such as rescue and relief work, and the like, and the mobile edge network has physical link characteristics different from the Internet and mainly comprises the following points: a) high transmission delay, b) high bit error rate, c) intermittent connection, d) asymmetric links, which causes unstable intermittent network links in technical links such as information transmission and data transmission due to changes of external environment. Meanwhile, as the service node is in a maneuvering state, the service node is subject to environmental interference at any time and is influenced by large-area loads of information processing and calculation tasks, real-time decision on the system state of the information service node is difficult to ensure, so that the current task requirements are met.
Therefore, the invention provides a mobile edge information service network system in an intermittent network environment, which realizes the cooperative work among all nodes in the mobile edge information service network.
Disclosure of Invention
The invention is realized by adopting the following technical scheme:
a mobile edge information service network comprising a plurality of nodes interconnected by wireless communication, wherein: the nodes comprise general nodes and main nodes, each node is provided with a task scheduling management system, the task scheduling management system comprises a task manager, a task executor, a task information collector and a task scheduler, the task manager and the task executor can operate on each node, and the task information collector and the task scheduler only operate on the main node.
The network, wherein: the task manager is used for maintaining a task list, comprising an input variable part, an output part and meta information of the task, and is also used for maintaining the resource use condition of the node.
The network, wherein: the task manager provides an adding, deleting and checking interface of the task and an updating interface of the resource use condition, and supports event subscription, and actively pushes the task meta-information or the resource use condition to subscribers when the meta-information of the task changes and the resource use condition changes.
The network, wherein: the identity of the node in the network is determined by the resource manager of the node, and after the resource manager determines the identity of the own main node, the information collector and the scheduler are started, and once the identity of the main node is lost, the information collector and the scheduler which are started are stopped immediately.
The network, wherein: when the network is initially built, all nodes are general nodes, the resource manager regularly broadcasts heartbeat information added with the weight of the node where the resource manager is located, and once the resource manager receives the heartbeat with the weight larger than that of the resource manager, the state of the general nodes is maintained for a period of time; once the resource manager has broadcast a predetermined number of heartbeat packets as a general node and has not received a heartbeat signal with a weight greater than that of itself during this period, it changes its identity to the master node.
The network, wherein: the task executor is used for executing tasks, updating the use condition of resources and reporting the execution result of the tasks; the task executor subscribes all tasks scheduled to the node where the task executor is located to the task manager, when a new task to be executed is detected, the task is submitted to a service framework of the node to execute the task, the execution starting time and the estimated time consumption of the task are updated by utilizing an interface of the task manager, and the completion state and the output of the task are updated after the task execution is finished.
The network, wherein: the task information collector is used for subscribing and collecting the computing resources and task information of each node from each node, periodically broadcasting a data packet to inquire the on-line task manager, actively replying after the task manager receives the data packet, and acquiring the node information from the task manager and subscribing all tasks after the task information collector senses that the task manager is on line, so that a global resource and task view is maintained.
The network, wherein: the task scheduler is used for acquiring global resources, unfinished task information and dynamic change events thereof from the task information collector and designating executors for the tasks.
The network, wherein: under the intermittent weak connection network condition, if the network connection of the main node is temporarily interrupted, the main node is reselected by the residual resource manager in the network, the new main node starts the information collector and the scheduler, and the new information collector acquires the full resource information and the task meta information from the resource manager of each general node.
The network, wherein: after the main node is determined, the main node selects one other node as a secondary main node, and when the other node becomes the secondary main node, the information collector is started, but the scheduler is not started; when the pressing of the master node disappears for a preset time, the auxiliary master node immediately starts broadcasting the heartbeat, simultaneously starts the scheduler to take over the network, and after the master node returns, the pressing of the master node is restored, and the auxiliary master node will close the scheduler again.
The network, wherein: when the common node and the auxiliary master node are pressed by the master node, the common node and the auxiliary master node report own weight to the master node periodically, and the master node determines the auxiliary master node by the received weight report; the secondary master node will become the master node after having taken over the network and broadcast a predetermined number of heartbeat packets.
The network, wherein: the task scheduler reschedules tasks executed by the task manager originally scheduled to a certain node to other task managers after finding out that the network connection of the certain node is interrupted for a certain period of time, and the task owners push the tasks to new executors after confirming the change of the task executors.
The network, wherein if the offline task manager resumes the network connection, the task information collector can sense the event and pass the event to the task scheduler, so that the task scheduler can sense the event and find that the same task is executed under a plurality of task managers, and the task scheduler keeps the task with the earliest estimated end time and deletes all other tasks.
Drawings
FIG. 1 is a schematic diagram of a mobile edge information service network architecture;
FIG. 2 is a task manager collaboration diagram;
FIG. 3 is a schematic diagram of a task executor in cooperation with a task manager;
FIG. 4 is a schematic diagram of a task information collector in cooperation with a task manager;
FIG. 5 is a schematic diagram of a task scheduler and task information collector in cooperation with a task manager;
FIG. 6 is a diagram of task manager offline and offline behavior—actor offline and offline.
Detailed Description
As shown in fig. 1, the mobile edge information service network includes a plurality of mobile edge information service nodes, which typically appear as computers (servers) or the like, that are coupled to each other by wireless communication. The maneuvering edge information service node is provided with a task scheduling management system, and the task scheduling management system comprises four components, namely a task manager, a task executor, a task information collector and a task scheduler. External participants to the system include task submitters and administrators who view the state of the system. Four components are deployed at each mobile edge information service node, a task manager and a task executor are operated at each mobile edge service node, and a task information collector and a task scheduler are operated only on the mobile edge service node (mobile service center or master node) serving as a master node.
The task manager is configured to maintain a task list. The main attributes of the task are shown in table 1 below, and the other attributes in table 1 except the input variable part and the output two attributes are collectively referred to as meta information of the task. The task manager is also used for maintaining the resource usage of the mobile edge service node.
TABLE 1 task principal Properties
The task manager provides an add-drop-modify-check interface for tasks and an update interface for resource usage. The task manager supports event subscription, and actively pushes the task meta-information or the resource use condition to the subscriber when the meta-information of the task changes and the resource use condition changes. When subscribing to an event, the scope of the task is defined by the set conditions (e.g., only subscribing to information about a certain task) is supported.
As shown in fig. 2, the task manager is deployed with an instance at each mobile edge service node, where the task manager is carried by a terminal device, such as a car, that serves as the mobile edge service node. The application on the mobile service node connected with the main node can submit the task to the task manager instance through the interface provided by the task manager, then order the task, acquire the output of the task at the first time (namely, the sub-node generates the task, submits the task to the main node, the main node performs decomposition scheduling on the task, and then the sub-node acquires the output from the main node).
After the task's executor determines, if the master node itself is not the task's executor, then an attempt will be made to push the task to the executor, i.e., other motorized edge service nodes.
In a mobile edge information service network, a master node is unique and is selected from nodes in the network, other nodes can be called general nodes, each node is provided with a resource manager, the identity of the node in the network can be determined through the resource manager, after the resource manager determines the identity of the master node, the information collector and the scheduler can be started, and once the identity of the master node is lost, the information collector and the scheduler which are started can be stopped immediately.
Each resource manager determines its own weight value in a certain way, which may be a value set by an administrator in an auxiliary manner, or a value obtained by comprehensively considering its own position in the network topology, its own network connection quality, its own resource condition, and other factors. It is necessary to ensure that the weights of all resource managers do not repeat in the network.
The resource manager determines the master node through a heartbeat mechanism. The initial node is a common node, and the resource manager can broadcast heartbeat signals, and the heartbeat signals are attached with own weights. Once the resource manager receives the heartbeat with a larger weight than the heartbeat, the resource manager keeps the state of being suppressed for a period of time, namely, keeps the state of the general node. The master node may also be suppressed and once the master node is suppressed, it becomes a generic node. The master node will broadcast the heartbeat periodically to maintain its authority and suppress other general nodes. Once the resource manager leaves the hold-down state (i.e., no heartbeat information with a weight greater than itself is received within a predetermined time), it is considered that there is no master node in the current network and the heartbeat is started to be broadcast periodically. Once the resource manager as a general node broadcasts a certain number of heartbeat packets and does not receive heartbeat signals with a weight larger than that of the resource manager, the resource manager changes the identity of the resource manager into a master node.
The task executor is used for executing tasks, updating the use condition of resources and reporting the execution result of the tasks. As shown in fig. 3, the task executor subscribes all tasks scheduled to the mobile edge service node where the task executor locates to the task manager, when a new task to be executed is detected, the task is submitted to the service framework of the mobile edge service node, the mobile edge service node executes the task, the execution start time and the expected time consumption of the task are updated by using the interface of the task manager, and the completion state and the output of the task are updated after the execution of the task is finished.
The task information collector is used for subscribing and collecting computing resources and task information of the service nodes from each maneuvering edge service node: as shown in FIG. 4, the task information collector periodically broadcasts a data packet to inquire the on-line task manager, and the task manager can actively reply after receiving the data packet, so that the task information collector can sense the on-line and off-line events of the task manager.
After a task information collector senses that a task manager is on line, the node information is acquired from the task manager and all tasks are subscribed, so that a global resource and task view is maintained for a system manager to browse and view, and data support is provided for a task scheduler to make decisions.
The task scheduler is used for acquiring global resources, unfinished task information and dynamic change events thereof from the task information collector and designating executors for the tasks (the tasks can be decomposed and scheduled to be executed by a plurality of nodes). When the task has no executor or the executor of the task is not on line, after the task generates the scheduling requirement, the task scheduler calculates the executor suitable for the task through a scheduling algorithm, and changes the attribute of the executor of the task through the interface of the task owner, as shown in fig. 5.
The scheduling algorithm is specifically as follows: the execution progress of each task is proportional to the number of unit resources (e.g., computational power of a CPU) occupied by the task, P1 is the rate at which the unit resources operate the task in a unit time, M is the number of resources occupied by the task, t1 represents the time that has been executed, and the execution progress P of the task (assuming that the total task amount is 1, and P is normalized, P is equal to or greater than 0 and equal to or less than 1) is represented by formula (1):
P＝P 1 Mt 1 (1)
if the job is to be completed at the time of the deadline t2 and the required resource amount is Q, then the job is of the formula (2)
P 1 (M+Q)(t 2 -t 1 )＝1-P (2)
Can be pushed out according to the formula (1) and the formula (3)
We can obtain formula (4) from formulas (2) and (3)
And defining the resource demand in the current unit time as R, and obtaining the formula (5).
The job can be dynamically scheduled according to the unit resource demand R of each job, and the node meeting the resource demand is scheduled by a task scheduler according to the information in the current network collected by the task information collector. Specifically, the task scheduler can obtain the calculation power which can be provided by each node in the current network in the idle time through the information in the current network collected by the task information collector, if a certain node singly meets the requirement of R, the task is scheduled to a single node, if the single node can not meet the requirement, the task is distributed to a plurality of nodes to finish, and the sum of the calculation power of the selected nodes in the idle time in unit time is larger than or equal to R.
The main node selection mode under the intermittent weak connection network condition is as follows: if the network connection of the master node is temporarily interrupted, this will result in the remaining resource manager in the network reselecting the master node. The new master node will newly start the information collector and task scheduler. The new information collector, because it has just started, will acquire the full amount of resource information and task meta information from the resource manager of each general node. Waiting for the old master node network communication to resume, the new master node may give way to the old master node again, and the newly started information collector and task scheduler will stop again. If the network is frequently connected and disconnected again, the process is repeated, so that a lot of bandwidth resources are wasted, and because the main node needs to consume a lot of time, a considerable part of time is in a state that the whole system does not have the main node in the process, and a new task may not be scheduled in time.
In order to reduce the negative impact of the master node being offline. The master node may be enabled to select one other node as a secondary master node. The other nodes, when becoming secondary master nodes, also activate the information collector as the master node, but do not activate the scheduler. When the pressing of the master node disappears for a preset time, the auxiliary master node immediately starts broadcasting the heartbeat, and simultaneously starts the scheduler to take over the work of the master node (but not switch the identity as the master node), so that the process of reselecting the master node is omitted. And the task and resource change event of the usual receiving resource manager are utilized to replace the whole data acquired when accidents happen. After the primary node comes back, the throttle of the primary node resumes and the secondary primary node will re-close the scheduler. When the common node and the auxiliary master node are pressed by the master node, the weight of the common node and the auxiliary master node can be reported to the master node periodically. The master node decides who the secondary master node is through the received weight report. The reporting interval and validity period of the weights may be relatively long in order to avoid that instability of the secondary master network results in frequent handovers of secondary master nodes. The secondary master node will become the master node after taking over the network and broadcasting a certain number of heartbeat packets, which may be relatively large, so as to avoid temporary off-line of the master node leading to a handover of the master node.
In the weak connection condition, when the communication network connection of a certain general node is interrupted (whether the communication network connection is disconnected temporarily or disconnected for a long time cannot be judged), the task information collector can sense the event and transmit the event to the task scheduler.
As shown in fig. 6, when a communication connection of a certain node is found to be interrupted for a certain period of time, the task scheduler will reschedule the task executed by the task manager originally scheduled to the node with the interrupted connection to other task managers. After confirming the change of the task executor, the task owner pushes the task to a new executor. At this time, if the offline task manager returns again, the task information collector may perceive the event and pass the event to the task scheduler, so that the task scheduler may perceive the event and find that the same task is executed under multiple task managers, at which time the task scheduler may only keep the expected end time earliest and delete all others.
If the owner (i.e. the master node) of the task goes offline, the mobile service node can also find the task information collector through the task manager of other mobile service centers, thereby finding the task meta-information and the task output recorded at the executor of the task.
Network splitting is the same as the initial performance of a weakly connected network environment, and the network connection between mobile service centers is disconnected, but for the scene of network splitting, the environment change is finished, namely the disconnected network connection is not recovered in a short time, and the scene of the weakly connected network environment, and the environment change also comprises subsequent connection recovery.
In the network splitting environment, each split part finally has its own master node, which may be the previous master node, may be the previous slave master node after a period of time, or may be reselected. These master nodes will eventually reselect the secondary master nodes (depending on the newly perceived network composition, the secondary master nodes may not be selected because there are too few members). Tasks lost due to network splitting are also rescheduled by the scheduler and sent back to the task owner after execution
The split network can generate network merging under the condition that the network environment is well changed, the network merging can cause a plurality of main nodes and auxiliary main nodes to appear in the network at the same time, and one of the main nodes with the largest weight can quickly press other main nodes into auxiliary nodes due to the heartbeat rule, so that the main nodes become the only main nodes. And then the auxiliary main node can be reselected by the main node by utilizing an auxiliary main node selection mechanism, and the network merging has no influence on the submitted task because the task information is maintained in a scattered manner.
According to the invention, an information service network can be constructed aiming at the intermittent weak-connection maneuvering edge environment, so that the decomposition and scheduling requirements on complex tasks are met, and the smooth execution of the tasks is ensured.

Claims (3)

1. A mobile edge information service network comprising a plurality of nodes, the nodes being interconnected by wireless communication, characterized in that: the nodes comprise general nodes and main nodes, each node is provided with a task scheduling management system, the task scheduling management system comprises a task manager, a task executor, a task information collector and a task scheduler, the task manager and the task executor can operate in each node, and the task information collector and the task scheduler only operate in the main node; the task executor is used for executing tasks, updating the use condition of resources and reporting the execution result of the tasks; the task executor subscribes all tasks scheduled to the node where the task executor is located to the task manager, when a new task to be executed is detected, the task is submitted to a service framework of the node to execute the task, the execution starting time and the estimated time consumption of the task are updated by utilizing an interface of the task manager, and the completion state and the output of the task are updated after the task execution is finished.
2. The network of claim 1, wherein: the task manager is used for maintaining a task list, comprising an input variable part, an output part and meta information of the task, and is also used for maintaining the resource use condition of the node.
3. The network of claim 1, wherein: the task manager provides an adding, deleting and checking interface of the task and an updating interface of the resource use condition, and supports event subscription, and actively pushes the task meta-information or the resource use condition to subscribers when the meta-information of the task changes and the resource use condition changes.
CN201910758332.9A 2019-08-16 2019-08-16 Maneuvering edge information service network in intermittent network environment Active CN110502342B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910758332.9A CN110502342B (en) 2019-08-16 2019-08-16 Maneuvering edge information service network in intermittent network environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910758332.9A CN110502342B (en) 2019-08-16 2019-08-16 Maneuvering edge information service network in intermittent network environment

Publications (2)

Publication Number Publication Date
CN110502342A CN110502342A (en) 2019-11-26
CN110502342B true CN110502342B (en) 2023-07-18

Family

ID=68588139

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910758332.9A Active CN110502342B (en) 2019-08-16 2019-08-16 Maneuvering edge information service network in intermittent network environment

Country Status (1)

Country Link
CN (1) CN110502342B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111526179A (en) * 2020-03-30 2020-08-11 中科九度(北京)空间信息技术有限责任公司 Distributed node resource elastic expansion device and method
CN112910981B (en) * 2021-01-27 2022-07-26 联想(北京)有限公司 Control method and device

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6877107B2 (en) * 2001-07-05 2005-04-05 Softwired Ag Method for ensuring operation during node failures and network partitions in a clustered message passing server
CN103534982A (en) * 2013-04-09 2014-01-22 华为技术有限公司 Method, equipment and network virtualization system for protecting business reliability
CN104539681A (en) * 2014-12-23 2015-04-22 北京超图软件股份有限公司 Distributed GIS accelerating system and GIS service processing method
CN106059879A (en) * 2016-01-29 2016-10-26 深圳市禾望电气股份有限公司 Control system and control method thereof
CN109104346A (en) * 2017-06-21 2018-12-28 比亚迪股份有限公司 Train network data transmission method, system and its apparatus based on CANopen agreement
CN109104349A (en) * 2017-06-21 2018-12-28 比亚迪股份有限公司 Train network data transmission method, system and its apparatus based on CANopen agreement
US20190116128A1 (en) * 2017-10-18 2019-04-18 Futurewei Technologies, Inc. Dynamic allocation of edge computing resources in edge computing centers

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6877107B2 (en) * 2001-07-05 2005-04-05 Softwired Ag Method for ensuring operation during node failures and network partitions in a clustered message passing server
CN103534982A (en) * 2013-04-09 2014-01-22 华为技术有限公司 Method, equipment and network virtualization system for protecting business reliability
CN104539681A (en) * 2014-12-23 2015-04-22 北京超图软件股份有限公司 Distributed GIS accelerating system and GIS service processing method
CN106059879A (en) * 2016-01-29 2016-10-26 深圳市禾望电气股份有限公司 Control system and control method thereof
CN109104346A (en) * 2017-06-21 2018-12-28 比亚迪股份有限公司 Train network data transmission method, system and its apparatus based on CANopen agreement
CN109104349A (en) * 2017-06-21 2018-12-28 比亚迪股份有限公司 Train network data transmission method, system and its apparatus based on CANopen agreement
US20190116128A1 (en) * 2017-10-18 2019-04-18 Futurewei Technologies, Inc. Dynamic allocation of edge computing resources in edge computing centers

Also Published As

Publication number Publication date
CN110502342A (en) 2019-11-26

Similar Documents

Publication Publication Date Title
EP2519037B1 (en) Method and network for sharing sensor data among mobile terminals
CN110502342B (en) Maneuvering edge information service network in intermittent network environment
US20130067484A1 (en) Information processing apparatus, information processing method, recording medium and information processing system
CN100465901C (en) Network system, management computer, cluster management method, and computer program
US8095495B2 (en) Exchange of syncronization data and metadata
CN101771723A (en) Data synchronization method
CN102281329A (en) Resource scheduling method and system for platform as a service (Paas) cloud platform
CN102209047A (en) Instant messaging system and method for realizing information sharing
CN101931656A (en) ISP-friendly distributed service node selection and update method
CN103020810A (en) System and method for sharing schedule plans
CN102055644A (en) Method, device and system for load management in distributed directory service system
KR20110081867A (en) Always ready client/server data synchronization
CN103475742A (en) Method and system for determining master control node in cloud computing environment
CN103744735A (en) Method and device for scheduling multi-core resource
CN101267449B (en) A tree P2P system resource transmission method based on mobile agent mechanism
CN102209049B (en) Instant messaging system and method for user information processing
Zhu et al. Online scheduling of transmission and processing for aoi minimization with edge computing
CN102209051B (en) Instant messaging system and method for realizing user information sharing
CN100456705C (en) Monitoring system and method for Internet multimedia communication
CN102209050B (en) Instant messaging system and method for realizing user information sharing
Koldehofe Buffer management in probabilistic peer-to-peer communication protocols
CN101340423A (en) Multi-cluster job scheduling method based on element scheduling ring
CN102209048B (en) Instant messaging system and method for realizing information sharing for users
CN112686004B (en) Block chain-based single-document multi-interface editing method
CN114598662A (en) Message queue cluster federal management system and method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant