CN106790403A - Realize the method for mobile cloud computing halfpace and realize distributed method - Google Patents
Realize the method for mobile cloud computing halfpace and realize distributed method Download PDFInfo
- Publication number
- CN106790403A CN106790403A CN201611074053.3A CN201611074053A CN106790403A CN 106790403 A CN106790403 A CN 106790403A CN 201611074053 A CN201611074053 A CN 201611074053A CN 106790403 A CN106790403 A CN 106790403A
- Authority
- CN
- China
- Prior art keywords
- intelligent terminal
- cloud computing
- halfpace
- resource
- mobile cloud
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- 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]
- G06F9/5061—Partitioning or combining of resources
-
- 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
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0806—Configuration setting for initial configuration or provisioning, e.g. plug-and-play
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/562—Brokering proxy services
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer And Data Communications (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a kind of method for realizing mobile cloud computing halfpace and realize distributed method, on the one hand, the mobile cloud computing halfpace of the present invention is set up between Hadoop and android system, for ensureing the adaptation between Hadoop and android system and operation, so that Hadoop realizes the normal operation in android system, that is, cause that android system supports the operation of Hadoop.On the other hand, by the mobile cloud computing halfpace of the present invention, intelligent terminal is realized very easily to be set up, adds, removing the distributed system that wireless network/cable network is set up, and be achieve particularly and is set up distributed type assemblies using intelligent terminal that is more cheap and being convenient for changing.
Description
Technical field
The present invention relates to big data technology, espespecially a kind of method for realizing mobile cloud computing halfpace and realization are distributed
Method.
Background technology
Hadoop is a distributed system architecture developed by Apache funds club.User can not know about
In the case of distributed low-level details, distributed program is developed.Making full use of the power of cluster carries out high-speed computation and storage.
The characteristics of Hadoop distributed file systems (HDFS, Hadoop Distributed File System) have high fault tolerance,
And be designed for being deployed in (low-cost) PC server more cheap compared to minicomputer；And HDFS can provide height and handle up
Amount (high throughput) carrys out the data of access application, and being adapted to those has super large data set (large data
Set application program).HDFS relaxes the requirement of (relax) POSIX, can in the form of streaming access (streaming
Access) the data in file system.
The design that the framework of Hadoop is most crucial is exactly：HDFS and MapReduce.HDFS is deposited for the data of magnanimity are provided
Storage, and MapReduce provides calculating for the data of magnanimity.At present, Hadoop increase income freely, hardware is cheap, exploitation facility, because
This, leading position is occupied in the current in the market of big data so that the Hadoop market shares are incremented by a high speed year by year.
In fact, outside the cost of software, the financial cost that hardware is produced is more considerable.How to reduce hardware cost is
Each server manufacturer is considering and is seeking the problem of solution.
At present, the CPU (CPU, Central Processing Unit) of Intel (Intel) company is occupied
The monopoly position in market, causes the CPU of PC server to hold at high price.And the CPU processor price used by intelligent terminal
It is cheap.Both price differs 30 times or so, and power consumption differs 1000 times or so.And performance gap between the two, it is average to list
On core, Xeon only has 5 times or so of mobile terminal (such as mobile phone) processor.One collection of 100 intelligent terminals composition
Group's (such as mobile phone cluster), cost can be controlled at 1 to 2 ten thousand yuan, and physical size and a common PC server quite, calculate performance
It it is one 5-10 times of PC server, power consumption can be controlled 1/10 to the 1/20 of server.
At present, the android system based on intelligent terminal can not support the operation of Hadoop.In addition, existing intelligent terminal
Cluster is set up, in addition to being set up by wireless signal (mobile network, WIFI network), it is also possible to the hardware of intelligent terminal
Transformed, such as set up wire communication module.The mode of this component cluster, depends on the realization of hardware, realizes cumbersome
Complexity, and be difficult to be controlled cost.
The content of the invention
In order to solve the above-mentioned technical problem, the present invention provides a kind of method for realizing mobile cloud computing halfpace and realization
Distributed method, enables to android system to support the operation of Hadoop.
In order to reach the object of the invention, the invention provides a kind of method for realizing mobile cloud computing halfpace, including：
Carry out network configuration；According to the running configuration of the intelligent terminal setting Hadoop platform for adding distributed system；
Resource to calculating and store is managed, based on calculating and storage resource is implement resource integration treatment；
Information according to network configuration and the calculating after integrating and storage resource carry out cloud computing management；
The mobile cloud computing halfpace is set up between Hadoop platform and Android android system.
Alternatively, also include：For the Hadoop platform provides the Java storehouses that the android system lacks.
Alternatively, the network configuration includes：
When the mobile cloud computing halfpace starts for the first time, to install the intelligence of the mobile cloud computing halfpace
One fixing address of terminal distribution；
Connected mode according to the affiliated distributed system of the intelligent terminal obtains IP address.
Alternatively, it is described to be included according to the connected mode of intelligent terminal affiliated distributed system acquisition IP address：
When the connected mode is wired connection, the IP address of all intelligent terminals of the distributed system is searched, it is right
Maximum IP address processed after as the IP address；
When the connected mode is wireless connection, the IP address is divided automatically by wireless network dynamic host configuration protocol DHCP
Match somebody with somebody；
When the connected mode is for customization frame mode, the difference according to groove position obtains the IP address.
Alternatively, the described pair of resource for calculating and storing is managed, and is implement resource integration with storage resource based on calculating
Treatment includes：
When the affiliated distributed system of intelligent terminal for installing the mobile cloud computing halfpace receives write-in and calculates
During task, the resource behaviour in service in current distributed system is confirmed；
Resource behaviour in service according to confirming transfers the method for allocating tasks of the Hadoop platform：When storage size exceeds
Set the limit upper limit does not receive write operation, when computing capability does not receive calculating task beyond the setting limit upper limit；When depositing
When storage size is less than setting limit lower limit, preferential write-in, the upper limit until reaching setting limit；When computing capability is limited less than setting
During volume lower limit, preferentially receive calculating task, the upper limit until reaching setting limit；Storage size and meter in the distributed system
The intelligent terminal that calculation ability is between upper and lower bound is randomly assigned when obtaining task.
Alternatively, also include：In the distributed system described in the intelligent terminal of the installation mobile cloud computing halfpace
Intelligent terminal provide hardware health monitoring, and provide reparation operation.
Alternatively, after the hardware goes wrong, also include：Carry out reconnecting/reboot operation, preset times are reconnected/weighed
Open after still failing, point out to replace the hardware.
Alternatively, also include：When value range of the system journal more than setting, FIFO FIFO storages are carried out.
Present invention also offers a kind of device for realizing moving cloud computing halfpace, at least including memory and place
Reason device, wherein, be stored with following executable instruction in memory：
Carry out network configuration；According to the running configuration of the intelligent terminal setting Hadoop platform for adding distributed system；It is right
Calculate and the resource of storage is managed, based on calculating and storage resource is implement resource integration treatment；According to the letter of network configuration
Calculating and storage resource after breath and integration carry out cloud computing management；
The mobile cloud computing halfpace is set up between Hadoop platform and Android android system.
Distributed method is realized present invention also offers one kind, including：
Obtain and install mobile cloud computing halfpace and Hadoop platform to intelligent terminal；
Mobile cloud computing halfpace carries out data in the affiliated distributed system of the intelligent terminal in Hadoop platform
Resource migration and computing resource are integrated；
Mobile cloud computing halfpace receives the intelligent terminal in the distributed system from Hadoop platform feedback
Storage and computing resource situation, the management of computing that racks of going forward side by side.
Alternatively, when there is new intelligent terminal to access the distributed system, also include：
Storage and computing resource situation of the mobile cloud computing halfpace according to the new intelligent terminal for accessing, enter again
Row data resource migration and computing resource are integrated.
Alternatively, when there is new intelligent terminal to access the distributed system, also include：
New intelligent terminal is connected to distributed system place network, obtains and installs in the middle of the mobile cloud computing
Platform and Hadoop platform；
New intelligent terminal uses the mobile cloud computing halfpace installed, according to network where the distributed system
Connected mode configures Hadoop system network.
Alternatively, also specifically included including obtaining IP address according to the connected mode of the affiliated distributed system of intelligent terminal：
When the connected mode is wired connection, the IP address of all intelligent terminals of the distributed system is searched, it is right
Maximum IP address processed after as the IP address；
When the connected mode is wireless connection, the IP address is divided automatically by wireless network dynamic host configuration protocol DHCP
Match somebody with somebody；
When the connected mode is for customization frame mode, the difference according to groove position obtains the IP address.
Alternatively, the acquisition and after installing mobile cloud computing halfpace and Hadoop platform, it is described to be distributed
Before data resource migration and computing resource are integrated in formula system, also include：
The mobile cloud computing halfpace provides the Java storehouses that the android system of the intelligent terminal lacks, to institute
Stating the android system of intelligent terminal carries out hardware selection, loading；
The distributed system connected mode that the mobile cloud computing halfpace is returned according to intelligent terminal is the intelligence
Terminal distribution IP address；And write the configuration file of Hadoop platform.
Alternatively, it is described to carry out data resource migration and computing resource and integrate to include：
Storage size does not receive write operation beyond the intelligent terminal of the setting limit upper limit, and computing capability is beyond setting limit
The functional terminal of the upper limit does not receive calculating task；
When the storage size of intelligent terminal is less than setting limit lower limit, preferential write-in, until reaching the upper of setting limit
Limit；When the computing capability of intelligent terminal is less than setting limit lower limit, preferentially receive calculating task, until reaching setting limit
The upper limit；
When the intelligent terminal of storage size and computing capability between upper and lower bound obtains task, it is randomly assigned.
Alternatively, also include：The mobile cloud computing halfpace stores daily record using FIFO storage modes.
Alternatively, also include：The distribution that the mobile cloud computing halfpace feeds back according to the Hadoop platform
The heartbeat situation of the intelligent terminal of formula system access carries out the monitoring of hardware health status.
Alternatively, the mobile cloud computing halfpace is supplied to the external interface to carry out cloud computing management.
Alternatively, when thering is intelligent terminal to exit the distributed system in the distributed system, also include：
The mobile cloud computing halfpace is received and exits distributed system application from the Hadoop platform,
Data resource migration and computing resource in the distributed system are carried out in the Hadoop platform again to integrate；
After the intelligent terminal is exited successfully, the mobile cloud computing halfpace log.
The mobile cloud computing halfpace of the present invention is set up between Hadoop and android system, for ensureing Hadoop
Adaptation and operation and android system between so that even if Hadoop realizes the normal operation in android system
Obtain the operation that android system supports Hadoop.On the other hand, by the mobile cloud computing halfpace of the present invention, intelligence is eventually
End realizes very easily to be set up, adds, removing the distributed system that wireless network/cable network is set up, and is especially realized
Use intelligent terminal establishment distributed type assemblies that are more cheap and being convenient for changing.
Other features and advantages of the present invention will be illustrated in the following description, also, the partly change from specification
Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages can be by specification, rights
Specifically noted structure is realized and obtained in claim and accompanying drawing.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, this hair
Bright schematic description and description does not constitute inappropriate limitation of the present invention for explaining the present invention.In the accompanying drawings：
Fig. 1 is the composition structural representation of the mobile cloud computing halfpace of the present invention；
Fig. 2 is the operation principle schematic flow sheet of the mobile cloud computing halfpace of the present invention；
Fig. 3 is the schematic flow sheet that HDFS cluster of the present invention based on android system realizes data storage；
Fig. 4 is the schematic flow sheet that HDFS cluster of the present invention based on android system is realized calculating；
Fig. 5 is the composition structural representation of computing module in improved intelligent terminal of the invention；
Fig. 6 is the composition structural representation of memory module in improved intelligent terminal of the invention；
Fig. 7 is networking schematic diagram of the intelligent terminal of the present invention using WIFI connected modes；
Fig. 8 is the embodiment of management to intelligent terminal in the cluster of the present invention based on the connected mode shown in Fig. 7
Schematic flow sheet；
Fig. 9 is networking schematic diagram of the intelligent terminal of the present invention using mobile network's connected mode；
Figure 10 is the embodiment of management to intelligent terminal in the cluster of the present invention based on the connected mode shown in Fig. 9
Schematic flow sheet；
Figure 11 is networking schematic diagram of the intelligent terminal of the present invention using wired connection mode；
Figure 12 is the embodiment of management to intelligent terminal in the cluster of the present invention based on the connected mode shown in Figure 11
Schematic flow sheet；
Figure 13 is the top view of frame in present invention customization rack construction；
Figure 14 is frame side view in present invention customization rack construction；
Figure 15 is to realize the schematic diagram of hardware replacement in present invention customization rack construction；
Figure 16 is the schematic flow sheet of the embodiment of the mobile cloud computing halfpace application of the present invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention become more apparent, below in conjunction with accompanying drawing to the present invention
Embodiment be described in detail.It should be noted that in the case where not conflicting, in the embodiment and embodiment in the application
Feature can mutually be combined.
Fig. 1 is the composition structural representation of the mobile cloud computing halfpace of the present invention, as shown in figure 1, the mobile cloud of the present invention
The intermediate layer that halfpace is built upon between Hadoop platform and android system is calculated, for ensureing that Hadoop puts down
Adaptation and operation between platform and android system.At least include：Hardware drive module, Java storehouses supporting module, configuration mould
Block, calculating SRM module, cloud computing task scheduling modules；Wherein,
Hardware drive module, for loading hardware driving for operating system.The driving is mainly network controller service, than
Such as：When communication is for RJ45 interfaces, network controller turns RJ45 signal chips for USB；When communication is for fiber optic communication, network control
Device processed is photoelectric conversion chip.
Java storehouses supporting module, for providing the Java storehouses that android system lacks for Hadoop platform.Android systems
System supports a part of Java functions in itself, and Java storehouses supporting module here is used to supply the Java lacked in android system
Storehouse.
Configuration module, for carrying out network configuration；According to the intelligent terminal such as the intelligent terminal that add distributed system
Computing capability, storage size set the running configuration of Hadoop platform automatically.
Wherein, network configuration is specifically included：
Mobile cloud computing halfpace where configuration module itself, when starting first time, is the installation mobile cloud computing
The intelligent terminal of halfpace distributes a fixing address (being all the IP address during all machine initial start-ups)；
Network connect after determine the affiliated distributed system of the intelligent terminal connected mode it is such as wired, wirelessly or customization machine
Frame type etc., IP address is obtained according to connected mode：When for wired connection, the IP ground of all machines of current network is searched
Location, is then processed maximum IP address：In the 4th section+1 of maximum IP address, when being added to 253, then in IP address the
Three sections Jia 1, and the 4th section starts counting up from 1；When for wireless connection, IP address by wireless network DHCP (DHCP,
Dynamic Host Configuration Protocol) automatic distribution, it is not necessary to consider that IP address is distributed, it is only necessary to read
The IP address of distribution simultaneously writes Hadoop configuration files；When for customization frame mode, intelligent terminal is according to groove position
Difference obtains IP address configuration, so so that the new device configuration for replacing upper electricity is identical with the device configuration for going wrong, i.e.,
The IP address of same groove position can not be changed.
SRM module is calculated, on the basis of existing Hadoop is calculated and storage is controlled, being incorporated into meter
Calculate in mobile cloud computing halfpace where SRM module itself, the resource to calculating and store is managed, base
Integration treatment is carried out in calculating with storage resource.
Implement including：When distributed system receives external write-in and calculating task, current distributed system is first confirmed
The resource behaviour in service of machine, then transfers the method for allocating tasks of Hadoop platform in system：Storage has exceeded setting limit
The upper limit will not receive write operation, and to calculate will not receive calculating task beyond the setting limit upper limit；When storage is less than setting
To preferentially be write during limit lower limit, the upper limit until reaching setting limit；To preferentially be connect when calculating less than setting limit lower limit
By calculating task, the upper limit until reaching setting limit；Other machines being between upper and lower bound are carried out when obtaining task
It is randomly assigned.
Cloud computing task scheduling modules, for being supplied to external interface, the meter after information and integration according to network configuration
Calculation and storage resource carry out cloud computing management, and management includes but is not limited to startup, stopping, pause, timing etc..
Further, the mobile cloud computing halfpace of the present invention also includes：Health monitoring module,
For providing hardware health monitoring for intelligent terminal, and provide reparation operation.It should be noted that health monitoring mould
Block monitoring hardware anyway, can carry out reconnecting/reboot operation after hardware goes wrong, and after repeatedly reconnecting/restarting unsuccessfully, point out
Replace the hardware.Monitoring function can carry out monitoring hardware anyway using heartbeat detection mechanism, and temperature in use monitors to enter hardware
Row operation health alarm etc..
Further, the mobile cloud computing halfpace of the present invention also includes：Log processing module,
For when system journal is more than the value range for setting, the storage of daily record to carry out FIFO (FIFO, First in
First out) storage, i.e., daily record at a specified future date is first deleted, then store newest daily record.
Fig. 1 only lists the basic Hadoop components of the mobile cloud computing halfpace of the present invention, and other assemblies can be certainly
Row selection is installed, if including other assemblies, including protection domains that are how many, being not intended to limit the present invention, no longer gone to live in the household of one's in-laws on getting married here
State.
Correspondingly, the present invention also provides a kind of method for realizing mobile cloud computing halfpace, including：Will mobile cloud computing
Halfpace is set up between Hadoop platform and Android android system, is also included：
Carry out network configuration；According to the running configuration of the intelligent terminal setting Hadoop platform for adding distributed system；
Resource to calculating and store is managed, based on calculating and storage resource is implement resource integration treatment；
Information according to network configuration and the calculating after integrating and storage resource carry out cloud computing management.
The inventive method also includes：For Hadoop platform provides the Java storehouses that android system lacks.
Alternatively, the network configuration for carrying out includes：
When the mobile cloud computing halfpace set up starts for the first time, to install the mobile cloud computing halfpace
Intelligent terminal distributes a fixing address；Connected mode according to the affiliated distributed system of the intelligent terminal obtains IP address.Its
In,
Connected mode according to the affiliated distributed system of intelligent terminal obtains IP address to be included：
When connected mode is wired connection, the IP address of all intelligent terminals of distributed system is searched, to maximum
IP address processed after as the IP address；
When connected mode is wireless connection, IP address is distributed automatically by wireless network DHCP；
When connected mode is for customization frame mode, the difference according to groove position obtains the IP address.
Alternatively, the resource for calculating and store is managed, based on calculating and storage resource is implement resource integration treatment
Including：
When the affiliated distributed system of intelligent terminal for installing mobile cloud computing halfpace receives write-in and calculating task
When, confirm the resource behaviour in service in current distributed system；
Resource behaviour in service according to confirming transfers the method for allocating tasks of Hadoop platform：When storage size is beyond setting
The limit upper limit does not receive write operation, when computing capability does not receive calculating task beyond the setting limit upper limit；When storage is big
During the small limit lower limit less than setting, preferential write-in, the upper limit until reaching setting limit；When computing capability is less than under setting limit
In limited time, calculating task is preferentially received, the upper limit until reaching setting limit；Storage size and calculating energy in the distributed system
The intelligent terminal that power is between upper and lower bound is randomly assigned when obtaining task.
Alternatively, the present invention realizes that the method for mobile cloud computing halfpace also includes：
Intelligent terminal in the distributed system described in the intelligent terminal of the mobile cloud computing halfpace of installation provides hardware
Health monitoring, and reparation operation is provided.
After hardware goes wrong, also include：Carry out reconnecting/reboot operation, preset times are reconnected/restarted after still failing,
The hardware is replaced in prompting.
Alternatively, the present invention realizes that the method for mobile cloud computing halfpace also includes：When system journal is more than setting
During value range, FIFO storages are carried out.
The present invention also provides a kind of device for realizing moving cloud computing halfpace, at least including memory and treatment
Device, wherein,
Be stored with following executable instruction in memory：Mobile cloud computing halfpace is set up in Hadoop platform and peace
Between tall and erect android system；Carry out network configuration；According to the fortune of the intelligent terminal setting Hadoop platform for adding distributed system
Row configuration；Resource to calculating and store is managed, based on calculating and storage resource is implement resource integration treatment；According to network
Calculating and storage resource after the information of configuration and integration carry out cloud computing management.
The mobile cloud computing halfpace set up using the present invention, realizes that distributed method includes：
Obtain and install mobile cloud computing halfpace and Hadoop platform to intelligent terminal；
Mobile cloud computing halfpace carries out data in the affiliated distributed system of the intelligent terminal in Hadoop platform
Resource migration and computing resource are integrated；
Mobile cloud computing halfpace receives the intelligent terminal in the distributed system from Hadoop platform feedback
Storage and computing resource situation, the management of computing that racks of going forward side by side.
With reference to an intelligent terminal after the mobile cloud computing halfpace shown in Fig. 1 of the present invention is installed each module
As a example by workflow situation, the operation principle to the mobile cloud computing halfpace of the present invention is described in detail.Fig. 2 is the present invention
The operation principle schematic flow sheet of mobile cloud computing halfpace, as shown in Fig. 2 including：
First, intelligent terminal is obtained and installs the present invention mobile cloud computing halfpace and Hadoop platform；The present invention is moved
The Java storehouses supporting module of dynamic cloud computing halfpace provides the Java storehouses that the android system of intelligent terminal lacks；The present invention
The Hardware drive module of mobile cloud computing halfpace carries out hardware selection, loading to the android system of intelligent terminal；This hair
The configuration module of bright mobile cloud computing halfpace is that intelligent terminal distributes an IP address：Intelligent terminal can be mobile to the present invention
The configuration module of cloud computing halfpace returns to distributed system connection type；The configuration of the mobile cloud computing halfpace of the present invention
Module carries out the configuration of corresponding IP address according to the heterogeneous networks connected mode of intelligent terminal：When for wired connection, search and work as
The IP address of all machines of preceding network, is then processed maximum IP address：In the 4th section+1 of maximum IP address, when
When being added to 253, then in IP address the 3rd section Jia 1, the 4th section starts counting up from 1；When for wireless connection, IP address is by wireless network
DHCP is distributed automatically；When for customization frame mode when, according to groove position difference come obtain IP address configuration.Intelligent terminal reads and divides
The configuration file of the IP address write-in Hadoop platform matched somebody with somebody simultaneously comes into force；
Then, the calculating SRM module of the mobile cloud computing halfpace of the present invention is carried out in Hadoop platform
Data resource migration and computing resource are integrated in distributed system, and Hadoop platform is by the storage of intelligent terminal and computing resource
Situation feeds back to the calculating SRM module of the mobile cloud computing halfpace of the present invention；
Then, the calculating SRM module of present invention movement cloud computing halfpace is whole according to the new intelligence for accessing
The storage at end and computing resource situation, carry out data resource migration and computing resource is integrated：Storage has exceeded setting limit
The upper limit will not receive write operation, and to calculate will not receive calculating task beyond the setting limit upper limit；When storage is less than setting
To preferentially be write during limit lower limit, the upper limit until reaching setting limit；To preferentially be connect when calculating less than setting limit lower limit
By calculating task, the upper limit until reaching setting limit；Other machines being between upper and lower bound are carried out when obtaining task
It is randomly assigned；
Finally also have, for the storage of daily record, the log processing module of the mobile cloud computing halfpace of the present invention is carried out
FIFO is stored；The health monitoring module of the mobile cloud computing halfpace of the present invention carries out the monitoring of hardware health, Hadoop platform
Feed back the heartbeat situation of the intelligent terminal for accessing to detect hardware anyway, feed back the node temperature of the intelligent terminal for accessing to monitor
Operation health status；The cloud computing task scheduling modules of the mobile cloud computing halfpace of the present invention are supplied to external interface to enter to rack
Management of computing.
Subsequently, if distributed system is exited in intelligent terminal application, Hadoop platform can be in the mobile cloud computing of the present invention
Between the calculating SRM module application of platform exit distributed system, the calculating of the mobile cloud computing halfpace of the present invention
SRM module can again carry out data resource migration and computing resource in distributed system in Hadoop platform
Integrate, and after exiting successfully, the log in the log processing module of the mobile cloud computing halfpace of the present invention.
In sum, in being the mobile cloud computing shown in Fig. 1 by Hadoop platform of the present invention based on android system
Between platform, on the one hand, so that Hadoop platform realizes normal fortune in android system and causes android system branch
The operation of Hadoop platform is held.Further, there is provided Hadoop platform run when software and hardware health monitoring so that
Ensure that and the reparation to software or the replacement to hardware are triggered when damaging.On the other hand, by the present invention based on Android systems
The Hadoop platform platform of system is the mobile cloud computing halfpace shown in Fig. 1, intelligent terminal realize very easily set up,
Add, remove the distributed distributed system that wireless network/cable network is set up.It is achieved thereby that using more cheap and convenient
The intelligent terminal of replacing sets up distributed system.
The embodiment of the present invention carries out citing description by mobile phone terminal of intelligent terminal, but, intelligent terminal of the present invention includes
But it is not restricted to：Mobile phone terminal, PAD, the Set Top Box of android system are run, and operation all kinds of of android system have
Line, wireless handheld terminal etc..
For convenience of describe, in the embodiment of the present invention in a distributed manner system be cluster as a example by be described, but be not used to limit
Determine protection scope of the present invention.
Need illustratively, Hadoop platform is by the mobile cloud computing halfpace of the present invention to bottom android system
Calling each application interface (API) function carries out the tasks such as reading and writing data, calculating.The encapsulation of Android api interfaces include but not
It is limited to：
For the read-write of intelligent terminal internal storage data, can include：
getFileDir()：For obtaining positional information of the Hadoop platform in intelligent terminal data stored in memory, such as：/
data/data/Hadoop/files；
getCacheDir()：For obtaining positional information of the Hadoop platform in intelligent terminal memory cache data, such as：/
data/data/Hadoop/cache；
openFileInput(String name)：For direct access/data/data/Hadoop/files/name texts
The input stream information of part；
openFileOutput(String name,int mode)：For direct access/data/data/Hadoop/
The output stream of files/name files, wherein, mode is authority when writing file；
Context.MODE_PRIVATE：Be privately owned pattern (or being default mode) can only be employed itself and it is same
The intelligent terminal of group is accessed；The content covering original content of write-in；
Context.MODE_APPEND：To add pattern (or being privately owned pattern), itself and same a group can only be employed
The intelligent terminal of group is accessed；Content is added if file is present, if file does not exist with regard to new files and writes content；
Context.MODE_WORLD_READABLE：For all intelligent terminals can read right in group；
Context.MODE_WORLD_WRITEABLE：For all intelligent terminals can write permission in group；
For SDcard reading and writing datas, can include：
getExternalStorageDirectory()：SDcard for obtaining intelligent terminal where Hadoop platform
Positional information, such as/storage/sdcard；
getExternalStorageState()：For obtaining working as the SDcard of intelligent terminal where Hadoop platform
Preceding state, the more commonly used should be MEDIA_MOUNTED；
FileInputStream (), for reading file；
BufferReader (), for reading file；
HttpConnection (), for the stream of reading to be saved as into String data；
FileOutputStream (), for writing file；
BufferedWriter (), for writing file；
In addition, Hadoop calculate using the adding of Java storehouses, subtract, multiplication and division arithmetic.
In the present invention, the HDFS clusters based on android system have two class nodes and with " manager-worker " pattern
Operation：One master (Master) node (NameNode) and multiple as manager is used as worker from (Slave) node
(DataNode).Each NameNode and DataNode one intelligent terminal of each correspondence.
Wherein, NameNode, is mainly responsible for HDFS file system, specifically including namespace management, block
(block) manage；DataNode, is mainly used in data storage file.Here, HDFS can be by a file division into one by one
Block, these block be potentially stored on a DataNode or multiple DataNode on.DataNode is responsible for reality
Bottom document read-write, if client (Client) program initiated read HDFS on file order, then, first can
These files are divided into several block, then NameNode will inform which Client these block data are stored in
On DataNode, so, Client is directly interacted with DataNode.
Storage and the calculating process of HDFS cluster of the present invention based on android system is described in detail below.
Fig. 3 is the schematic flow sheet that HDFS cluster of the present invention based on android system realizes data storage, with file
Pass to as a example by distributed type assemblies are stored, here, NameNode is responsible for the metadata of storage All Files on HDFS,
The request of client can be confirmed, and record the back end i.e. set of DataNode of the name and storage this document of file.
And in file allocation table of the information Store for recording these in internal memory.In the present embodiment, it is assumed that client sends one
Ask to NameNode, show for " swim.txt " file to be written to HDFS, as shown in figure 3, including：
Step 1：Client sends out message to NameNode, shows to write " swim.txt " file；
Step 2：NameNode returns to message to client, notifies that " swim.txt " file is write DataNode by client
A, DataNode B and DataNode D, and directly contact with DataNode B；
Step 3：Client hair message gives DataNode B, indicates DataNode B to preserve a " swim.txt " file,
And a copy of transmission gives DataNode A and DataNode D respectively；
Step 4：DataNode B hair message gives DataNode A, indicates DataNode A to preserve a " swim.txt " text
Part, and send a copy and give DataNode D；
Step 5：DataNode A hair message gives DataNode D, indicates DataNode D to preserve a " swim.txt " text
Part；
Step 6：DataNode D hair confirmation messages give DataNode A；
Step 7：DataNode A hair confirmation messages give DataNode B；
Step 8：DataNode B send out confirmation message to client, represent the write-in for completing " swim.txt " file.
So, a " swim.txt " file be just stored in distributed type assemblies DataNode A, DataNode B and
On tri- back end of DataNode D (i.e. intelligent terminal).
Fig. 4 is the schematic flow sheet that HDFS cluster of the present invention based on android system is realized calculating, as shown in figure 4,
In HDFS based on Android, each MapReduce task is both initialized to a Job, and each Job can be divided into two again
The stage of kind：Mapping (Map) stage and reduction (Reduce) stage.The two stages are respectively with two function representations, i.e. Map functions
With Reduce functions.Wherein, Map functions receive ＜ a key, value>The input of form, then equally produces a ＜
key,value>Output in the middle of form；Reduce functions receive one such as ＜ key, (list of values)>Form it is defeated
Enter, then this value set is processed, each Reduce produces 0 or 1 output, and the output of Reduce is also ＜
key,value>Form.
Typically, common intelligent terminal makes after mobile cloud computing halfpace of the invention and Hadoop platform is installed
With the wireless connection of stabilization, Hadoop platform is carried out with postponing by mobile cloud computing halfpace of the invention, you can enter
The use of each component on row Hadoop.Such as in an office, everybody intelligent terminal can be focused on a cluster
It is interior, realize that computing resource and storage resource are shared.So, the amount of calculation that the intelligent terminal of people cannot be accomplished, by more than ten
Personal intelligent terminal can just set up cluster, and complete originally can not on mobile cloud computing halfpace paper of the invention
The amount of calculation of completion.Wireless connection includes again：WIFI network, two kinds of cluster teaming methods of mobile network (such as 2G, 3G, 4G).
In order to ensure reliability, the enhancing communication speed of cluster, it is possible to use wired connection mode.Using wired connection side
The cluster hardware source of formula can be the intelligent terminal of customization, it is also possible to carry out changing for wire communication using existing intelligent terminal
Make.Transformation to existing intelligent terminal mainly includes two parts：On the one hand it is that USB interface is transformed, by USB Type
Communication is converted to twisted-pair feeder or optical fiber cable communication；On the other hand, memory module at least includes ROM, SD card and interface.Storage
Module damage can be replacement.Specifically,
Fig. 5 is the composition structural representation of computing module in improved intelligent terminal of the invention, as shown in figure 5, at least
Including：Central processor unit (CPU, Central Processing Unit) chipset, random access memory (RAM, Random-
Access Memory), network controller.Externally include power supply, network interface, 3 interfaces of memory interface.
Wherein, USB interface can be included but is not limited to：USB2.0 or USB3.0, twisted-pair feeder and optical fiber cable are corresponded to respectively to be made
With.Therefore, two kinds of different types for USB use different communication plans.Network controller turns network interface card using common USB
Scheme, if optical fiber adds photoelectric conversion chip again.Network controller in Fig. 5 is a signal conversion chip,
Such as：USB2.0 turns RJ45 signal chips such as AX88772B；USB 3.0-to-Gigabit Ethernet control chips are such as
RTL8153 etc..
In Fig. 5, secure data (SD, Secure Digital) card interface and android system custom-built system mirror image are read-only
Memory (ROM, Read Only Memory Image) interface merges into memory interface, the simply change of hardware outward appearance, actual
Upper two interfaces or self-existent.
Fig. 6 is the composition structural representation of memory module in improved intelligent terminal of the invention, as shown in fig. 6, at least
Including built-in two components：ROM and SD card, are externally shown as a memory interface.The module damage can be replaced.
Hadoop is widely used big data platform, be currently based on Hadoop can be using this hair with embodiment
Bright offer and technical scheme is transferred in the intelligent terminal for being provided with android system by PC server.Below according to intelligence
The heterogeneous networks connected mode of terminal, management of the cloud computing halfpace to intelligent terminal in the cluster is moved with reference to the present invention
Embodiment is described in detail.
First embodiment
Fig. 7 is networking schematic diagram of the intelligent terminal of the present invention using WIFI connected modes, and Fig. 8 is based on Fig. 7 institutes for the present invention
The schematic flow sheet of the embodiment of the management to intelligent terminal in the cluster of the connected mode shown, with reference to Fig. 7 and Fig. 8, this reality
Example is applied so that new node is as intelligent terminal as an example, included the step of intelligent terminal is added into cluster：
Step 800：The intelligent terminal of the WIFI network for being intended to add existing cluster is successfully connected WIFI network.
Step 801：The intelligent terminal of the WIFI network for being intended to add existing cluster is downloaded by itself android system and obtained
And mobile cloud computing halfpace and Hadoop platform are installed.
Step 802：Hadoop platform grid is configured using mobile cloud computing halfpace.Process is implemented as schemed
Shown in 2, repeat no more here.
Assuming that connection (Connection) parameter of connected mode is 1, WIFI connected modes, Connection ginsengs are expressed as
Number is 2, is expressed as mobile network's such as 2G, 3G, 4G connected mode, and Connection parameters are 3, are expressed as wired connection mode.
In the present embodiment, it is assumed that Connection parameters are 1, then move cloud computing halfpace according in the WIFI network
DHCP mechanism distribute IP address to the new node (intelligent terminal such as in step 800) for adding in the present embodiment；And will distribution
IP address write-in cluster configuration file in.
Step 803：Host node in existing cluster starts Hadoop clusters component and successfully starts up.
Step 804：Host node in existing cluster carries out data resource migration and computing resource in cluster and integrates
(balance)。
In the present embodiment, mobile cloud computing halfpace in host node according to the storage of new node (such as intelligent terminal) and
Computing resource situation, voluntarily carries out data resource migration and computing resource is integrated.From the point of view of citing, it is assumed that the storage of each node
The resource upper limit is the 80% of total storage resource and storage resource lower limit is total storage resource 5%；Assuming that the CPU of each node
The computing resource upper limit reaches 80% and CPU computing resources lower limit for cpu busy percentage reaches 5% for cpu busy percentage；Wherein, on
Limit, the setting of lower limit can be adjusted according to actual demand.
Mobile cloud computing halfpace is proceeded as follows according to the upper limit, lower limit that set：
When the storage of the node of the new addition has exceeded the storage resource upper limit, write operation will not be received；
When the calculating of the node of the new addition exceeds the CPU computing resource upper limits, calculating task will not be received；
When the storage of the node of the new addition is less than storage resource lower limit, will preferentially write, until reaching storage resource
The upper limit；
When the calculating of the node of the new addition is less than CPU computing resource lower limits, will preferentially receive calculating task, Zhi Daoda
To the CPU computing resource upper limits；
Other are in the situation between upper and lower bound, can be randomly assigned when obtaining task.
Step 805~step 806：Host node in existing cluster calls mobile cloud computing halfpace interface, submits to
Hadoop tasks are run.
Step 807：Each back end in cluster completes task computation, and task result is fed back into host node.
In the present embodiment, it is assumed that add the intelligent terminal application of existing cluster to exit in certain back end such as step 800
Cluster, comprises the following steps：
Step 808：The intelligent terminal exits cluster to the host node application in cluster.
Step 809：Host node carries out data resource migration:And computing resource is integrated.Implement such as step 804 institute
State, repeat no more here.
Step 810：After the completion of resource consolidation, agree to that the intelligent terminal exits cluster.
Second embodiment
Fig. 9 is networking schematic diagram of the intelligent terminal of the present invention using mobile network's connected mode, and Figure 10 is based on for the present invention
The schematic flow sheet of the embodiment of the management to intelligent terminal in the cluster of the connected mode shown in Fig. 9, with reference to Fig. 9 and Tu
10, the present embodiment so that new node is as intelligent terminal as an example, by intelligent terminal add cluster the step of include：
Step 1000：The intelligent terminal for being intended to add mobile network such as 2G, 3G, 4G of existing cluster etc. is successfully connected movement
Network.
Step 1001：The intelligent terminal of the mobile network for being intended to add existing cluster is downloaded by itself android system and obtained
Take and install mobile cloud computing halfpace and Hadoop platform.
Step 1002：Hadoop system network is configured using mobile cloud computing halfpace.Implement process such as Fig. 2 institutes
Show, repeat no more here.
Assuming that connection (Connection) parameter of connected mode is 1, WIFI connected modes, Connection ginsengs are expressed as
Number is 2, is expressed as mobile network's such as 2G, 3G, 4G connected mode, and Connection parameters are 3, are expressed as wired connection mode.
In the present embodiment, it is assumed that Connection parameters are 2, then move cloud computing halfpace according in the mobile network
DHCP mechanism distribute IP address to the new node (intelligent terminal such as in step 1000) for adding in the present embodiment；And will distribution
IP address write-in cluster configuration file in.
Step 1003：Host node in existing cluster starts Hadoop clusters component and successfully starts up.
Step 1004：Host node in existing cluster carries out data resource migration and computing resource in cluster and integrates
(balance)。
In the present embodiment, mobile cloud computing halfpace in host node according to the storage of new node (such as intelligent terminal) and
Computing resource situation, voluntarily carries out data resource migration and computing resource is integrated.From the point of view of citing, it is assumed that the storage of each node
The resource upper limit is the 80% of total storage resource and storage resource lower limit is total storage resource 5%；Assuming that the CPU of each node
The computing resource upper limit reaches 80% and CPU computing resources lower limit for cpu busy percentage reaches 5% for cpu busy percentage；Wherein, on
Limit, the setting of lower limit can be adjusted according to actual demand.
Mobile cloud computing halfpace is proceeded as follows according to the upper limit, lower limit that set：
When the storage of the node of the new addition has exceeded the storage resource upper limit, write operation will not be received；
When the calculating of the node of the new addition exceeds the CPU computing resource upper limits, calculating task will not be received；
When the storage of the node of the new addition is less than storage resource lower limit, will preferentially write, until reaching storage resource
The upper limit；
When the calculating of the node of the new addition is less than CPU computing resource lower limits, will preferentially receive calculating task, Zhi Daoda
To the CPU computing resource upper limits；
Other are in the situation between upper and lower bound, can be randomly assigned when obtaining task.
Further,
In view of transmission rate and the factor of flow rate, data compression is also further set in the present embodiment
(Compress) parameter, when Compress parameters are 1, represents that needs are compressed, and when Compress parameters are 0, represents
Need not be compressed.
When Connection parameters are 2, i.e., in the present embodiment by the way of mobile network, Compress parameters are automatic
1 is set to, needs to be compressed when representing data transfer, to obtain high transfer rate, so as to reduce flow rate；
When Connection parameters are 1 or 3, Compress parameters are automatically set as 0, and data transfer need not be pressed
Contracting, to mitigate CPU computing pressure.
Step 1005~step 1006：Host node in existing cluster calls mobile cloud computing halfpace interface, submits to
Hadoop tasks are run.
Step 1007：Each back end in cluster completes task computation, and task result is fed back into host node.
In the present embodiment, it is assumed that add the intelligent terminal application of existing cluster to exit in certain back end such as step 1000
Cluster, comprises the following steps：
Step 1008：The intelligent terminal exits cluster to the host node application in cluster.
Step 1009：Host node carries out data resource migration:And computing resource is integrated.Implement such as step 1004 institute
State, repeat no more here.
Step 1010：After the completion of resource consolidation, agree to that the intelligent terminal exits cluster.
3rd embodiment
Figure 11 is networking schematic diagram of the intelligent terminal of the present invention using wired connection mode, and Figure 12 is based on Figure 11 for the present invention
The schematic flow sheet of the embodiment of the management to intelligent terminal in the cluster of shown connected mode, with reference to Figure 11 and Figure 12,
The present embodiment so that new node is as intelligent terminal as an example, by intelligent terminal add cluster the step of include：
Step 1200：It is intended to add the intelligent terminal of existing cluster to be successfully connected in cluster using twisted-pair feeder or optical fiber
Interchanger.In the present embodiment, it is intended to add the intelligent terminal of existing cluster to use USB2.0 interfaces, therefore, using twisted pair line connection
The interchanger of cluster.
Step 1201：The intelligent terminal of the mobile network for being intended to add existing cluster is downloaded by itself android system and obtained
Take and install mobile cloud computing halfpace and Hadoop platform.
Step 1202：Hadoop system network is configured using mobile cloud computing halfpace.Implement process such as Fig. 2 institutes
Show, repeat no more here.
Assuming that connection (Connection) parameter of connected mode is 1, WIFI connected modes, Connection ginsengs are expressed as
Number is 2, is expressed as mobile network's such as 2G, 3G, 4G connected mode, and Connection parameters are 3, are expressed as wired connection mode.
In the present embodiment, it is assumed that Connection parameters are 3, then move cloud computing halfpace and search current cluster network
The IP address of all nodes, is then processed maximum IP address：In the 4th section+1 of maximum IP address, when being added to 253
When, then in IP address the 3rd section Jia 1, the 4th section starts counting up from 1；And in the IP address write-in cluster configuration file that will be distributed.
Step 1203：Host node in existing cluster starts Hadoop clusters component and successfully starts up.
Step 1204：Host node in existing cluster carries out data resource migration and computing resource in cluster and integrates
(balance)。
In the present embodiment, mobile cloud computing halfpace in host node according to the storage of new node (such as intelligent terminal) and
Computing resource situation, voluntarily carries out data resource migration and computing resource is integrated.From the point of view of citing, it is assumed that the storage of each node
The resource upper limit is the 80% of total storage resource and storage resource lower limit is total storage resource 5%；Assuming that the CPU of each node
The computing resource upper limit reaches 80% and CPU computing resources lower limit for cpu busy percentage reaches 5% for cpu busy percentage；Wherein, on
Limit, the setting of lower limit can be adjusted according to actual demand.
Mobile cloud computing halfpace is proceeded as follows according to the upper limit, lower limit that set：
When the storage of the node of the new addition has exceeded the storage resource upper limit, write operation will not be received；
When the calculating of the node of the new addition exceeds the CPU computing resource upper limits, calculating task will not be received；
When the storage of the node of the new addition is less than storage resource lower limit, will preferentially write, until reaching storage resource
The upper limit；
When the calculating of the node of the new addition is less than CPU computing resource lower limits, will preferentially receive calculating task, Zhi Daoda
To the CPU computing resource upper limits；
Other are in the situation between upper and lower bound, can be randomly assigned when obtaining task.
Step 1205~step 1206：Host node in existing cluster calls mobile cloud computing halfpace interface, submits to
Hadoop tasks are run.
Step 1207：Each back end in cluster completes task computation, and task result is fed back into host node.
In the present embodiment, it is assumed that add the intelligent terminal application of existing cluster to exit in certain back end such as step 1200
Cluster, comprises the following steps：
Step 1208：The intelligent terminal exits cluster to the host node application in cluster.
Step 1209：Host node carries out data resource migration:And computing resource is integrated.Implement such as step 1004 institute
State, repeat no more here.
Step 1210：After the completion of resource consolidation, agree to that the intelligent terminal exits cluster.
Fourth embodiment
The present embodiment is to determine that cluster models are the situation for customizing rack construction after network is connected.
First, frame install, on electricity, frame unification is by AC conversion for direct current for all devices provide power supply
Afterwards, carry out release image operation, will whole system shown in Fig. 1 packing be stored directly in ROM as mirror image.
Frame axis entrance where installing the hardware store to backup of system.It is initial to install since it is desired that installing
Number of devices is more, it is necessary to artificial manual operations hardware supply.Follow-up alternate device is preferred with storing 5-10 block devices automatically.
Identical flow is replaced by hardware installation to the position specified by with backup, and as shown in figure 13, Figure 13 is this
The top view of frame in invention customization rack construction, as shown in figure 13, calculates memory module and is installed on the annulus of frame.This reality
Apply in example, annulus is rotated with low speed.
On the one hand, the effect of the wind-cooling heat dissipating for bringing legacy equipment that slowly runs of annulus, on the other hand, needs in hardware
When replacing, the docking for realizing and calculating storage backup module passage that slowly runs of annulus, in order to the mould that frame will be replaced
Block is transported in place；Meanwhile, centrifugal force is capable of the hardware of automatic disassembling damage.
Then, hardware UNICOM frame switching equipment obtains network parameter automatically, adds distributed type assemblies.
The equipment of cluster is incorporated at first for host node, can be adjusted by the cloud computing task of network connection host node in outside
Degree platform performs distributed computing task.
Then, alternate device is added, follow-up timing detects the alternate device quantity of frame, so as to addition in time.Need simultaneously
Reclaim the hardware device of damage.
Figure 14 is frame side view in present invention customization rack construction, and as shown in figure 14, the annular groove of frame position layer can be with
It is Multi-layer design, the line with arrow illustrates the operating path of backup hardware.
When hardware replacement is carried out, as shown in figure 15, locking is opened in hardware groove position, damages hardware because centrifugal force departs from machine
Frame；The communicated plant running of backup hardware is in backup guide groove；Annulus gradually reduces speed to stopping operating, make backup guide groove and
Hardware needs the groove position docking replaced, then will backup hardware loading hardware groove by the operation of conveyer；Change standby hard
After part, annulus gradually replys original operation rotary speed.
5th embodiment
The present embodiment is realized with typical application scenarios to the specific works of the mobile cloud computing halfpace of the present invention and system
It is described.Assuming that due to one game because it is computationally intensive and produce data it is many, run very interim card on a mobile phone, at the same need
Artificial intelligence analysis and excavation are carried out based on these big datas, there is provided player's scene finds and player recommends etc., the present embodiment
Middle to assume in a room, the smart mobile phone for having respective hand-held 1 android system of 5 people tests use this large-scale trip
Play.Figure 16 is the schematic flow sheet of the embodiment of the mobile cloud computing halfpace application of the present invention, in the present embodiment, as a example by describe
It is convenient, only so that 2 equipment are the operation flow between Master host nodes and a Slave node as an example, Master and other
Operation flow between Slave nodes is repeated no more here as the operation flow shown in Figure 16.As shown in figure 16, including：
Step 1600：Installed on Master nodes and Slave nodes respectively mobile cloud computing halfpace, Hadoop,
Game application.
Step 1601：Relevant configuration is carried out to android system and Hadoop platform using mobile cloud computing halfpace
And come into force, generally comprise：
The mobile cloud computing halfpace of host node provides the Java storehouses that android system lacks for Hadoop platform；It is main
The mobile cloud computing halfpace of node carries out hardware selection, loading to android system；In the middle of the mobile cloud computing of host node
Platform distributes an IP address, and the heterogeneous networks connected mode according to feedback carries out the configuration of corresponding IP address；Intelligent terminal
Android system reads the IP address write-in Hadoop configuration files of distribution and comes into force；It is flat in the middle of the mobile cloud computing of host node
Platform carries out data resource migration and computing resource in cluster in Hadoop platform and integrates, and notifies other in cluster
Resource adjustment is carried out in Slave nodes.Implement as shown in Fig. 2 repeating no more here.
Step 1602：Master nodes and the game application of Slave nodal tests in each node such as Figure 16 take resource feelings
Condition, and feed back to respective Hadoop platform；The Hadoop platform of each Slave nodes receives game application and takes resource situation
Afterwards, Master nodes are fed back to.
Step 1603：Master nodes carry out data resource migration automatically according to the storage of each node and computing resource situation
And computing resource is integrated.
Step 1604：Game application begins to use, and is each Slave sections after Master nodes receive usage behavior and task
Point distribution is calculated and store tasks；Each Slave nodes are to Master node feeding backs result of calculation and storage result, Master nodes
Carry out result of calculation and storage result collects, and to game application feedback game application operation result and storage result.
Step 1605：During the calculating task and store tasks of game application are carried out, the mobile cloud computing of host node
Halfpace carries out the storage of daily record, is stored using FIFO modes；And/or, the mobile cloud computing halfpace of host node is carried out
The hardware health monitoring of each node.
Step 1606：Game application terminates, and Master nodes notify that each Slave nodes release computing resource and storage are provided
Source；Each Slave nodes are to Master node feeding backs computing resource and storage resource releasing result.
The above, preferred embodiments only of the invention are not intended to limit the scope of the present invention.It is all this
Within the spirit and principle of invention, any modification, equivalent substitution and improvements done etc. should be included in protection model of the invention
Within enclosing.
Claims (19)
1. a kind of method for realizing mobile cloud computing halfpace, it is characterised in that including：
Carry out network configuration；According to the running configuration of the intelligent terminal setting Hadoop platform for adding distributed system；
Resource to calculating and store is managed, based on calculating and storage resource is implement resource integration treatment；
Information according to network configuration and the calculating after integrating and storage resource carry out cloud computing management；
The mobile cloud computing halfpace is set up between Hadoop platform and Android android system.
2. method according to claim 1, it is characterised in that also include：For the Hadoop platform provides described
The Java storehouses that android system lacks.
3. method according to claim 1, it is characterised in that the network configuration includes：
When the mobile cloud computing halfpace starts for the first time, to install the intelligent terminal of the mobile cloud computing halfpace
One fixing address of distribution；
Connected mode according to the affiliated distributed system of the intelligent terminal obtains IP address.
4. method according to claim 3, it is characterised in that the connection according to the affiliated distributed system of intelligent terminal
Mode obtains IP address to be included：
When the connected mode is wired connection, the IP address of all intelligent terminals of the distributed system is searched, to maximum
IP address processed after as the IP address；
When the connected mode is wireless connection, the IP address is distributed automatically by wireless network dynamic host configuration protocol DHCP；
When the connected mode is for customization frame mode, the difference according to groove position obtains the IP address.
5. method according to claim 1, it is characterised in that described pair calculates and the resource of storage is managed, and is based on
Calculate and storage resource treatment of implemening resource integration includes：
When the affiliated distributed system of intelligent terminal for installing the mobile cloud computing halfpace receives write-in and calculating task
When, confirm the resource behaviour in service in current distributed system；
Resource behaviour in service according to confirming transfers the method for allocating tasks of the Hadoop platform：When storage size is beyond setting
The limit upper limit does not receive write operation, when computing capability does not receive calculating task beyond the setting limit upper limit；When storage is big
During the small limit lower limit less than setting, preferential write-in, the upper limit until reaching setting limit；When computing capability is less than under setting limit
In limited time, calculating task is preferentially received, the upper limit until reaching setting limit；Storage size and calculating energy in the distributed system
The intelligent terminal that power is between upper and lower bound is randomly assigned when obtaining task.
6. the method according to any one of Claims 1 to 5, it is characterised in that also include：To install the mobile cloud computing
Intelligent terminal in distributed system described in the intelligent terminal of halfpace provides hardware health monitoring, and provides reparation operation.
7. method according to claim 6, it is characterised in that after the hardware goes wrong, also include：Carry out weight
Company/reboot operation, preset times are reconnected/restarted after still failing, and point out to replace the hardware.
8. the method according to any one of Claims 1 to 5, it is characterised in that also include：When system journal is more than setting
During value range, FIFO FIFO storages are carried out.
9. a kind of device for realizing moving cloud computing halfpace, at least including memory and processor, wherein, memory
In be stored with following executable instruction：
Carry out network configuration；According to the running configuration of the intelligent terminal setting Hadoop platform for adding distributed system；To calculating
Resource with storage is managed, based on calculating and storage resource is implement resource integration treatment；Information according to network configuration and
Calculating and storage resource after integration carry out cloud computing management；
The mobile cloud computing halfpace is set up between Hadoop platform and Android android system.
10. one kind realizes distributed method, it is characterised in that including：
Obtain and install mobile cloud computing halfpace and Hadoop platform to intelligent terminal；
Mobile cloud computing halfpace carries out data resource in the affiliated distributed system of the intelligent terminal in Hadoop platform
Migration and computing resource are integrated；
Mobile cloud computing halfpace receives depositing for the intelligent terminal in the distributed system from Hadoop platform feedback
Storage and computing resource situation, the management of computing that racks of going forward side by side.
11. methods according to claim 10, it is characterised in that access the distributed system when there is new intelligent terminal
When, also include：
Storage and computing resource situation of the mobile cloud computing halfpace according to the new intelligent terminal for accessing, re-start number
Integrated according to resource migration and computing resource.
12. methods according to claim 11, it is characterised in that access the distributed system when there is new intelligent terminal
When, also include：
New intelligent terminal is connected to distributed system place network, obtains and install the mobile cloud computing halfpace
And Hadoop platform；
New intelligent terminal uses the mobile cloud computing halfpace installed, according to the connection of network where the distributed system
Mode configures Hadoop system network.
13. methods according to claim 12, it is characterised in that also including according to the affiliated distributed system of intelligent terminal
Connected mode obtains IP address, specifically includes：
When the connected mode is wired connection, the IP address of all intelligent terminals of the distributed system is searched, to maximum
IP address processed after as the IP address；
When the connected mode is wireless connection, the IP address is distributed automatically by wireless network dynamic host configuration protocol DHCP；
When the connected mode is for customization frame mode, the difference according to groove position obtains the IP address.
14. methods according to claim 10, it is characterised in that the acquisition and install mobile cloud computing halfpace and
It is described to carry out in distributed system data resource migration and before computing resource is integrated after Hadoop platform, also include：
The mobile cloud computing halfpace provides the Java storehouses that the android system of the intelligent terminal lacks, to the intelligence
The android system of energy terminal carries out hardware selection, loading；
The distributed system connected mode that the mobile cloud computing halfpace is returned according to intelligent terminal is the intelligent terminal
Distribution IP address；And write the configuration file of Hadoop platform.
15. method according to claim 10 or 11, it is characterised in that described to carry out data resource migration and calculate money
Source is integrated to be included：
Storage size does not receive write operation beyond the intelligent terminal of the setting limit upper limit, and computing capability is beyond the setting limit upper limit
Functional terminal do not receive calculating task；
When the storage size of intelligent terminal is less than setting limit lower limit, preferential write-in, the upper limit until reaching setting limit；When
When the computing capability of intelligent terminal is less than setting limit lower limit, preferentially receive calculating task, the upper limit until reaching setting limit；
When the intelligent terminal of storage size and computing capability between upper and lower bound obtains task, it is randomly assigned.
16. method according to claim 10,11 or 12, it is characterised in that also include：It is flat in the middle of the mobile cloud computing
Platform stores daily record using FIFO storage modes.
17. diversity method according to claim 10,11 or 12, it is characterised in that also include：In the middle of the mobile cloud computing
The heartbeat situation of the intelligent terminal that the distributed system that platform feeds back according to the Hadoop platform is accessed carries out hardware and is good for
The monitoring of health situation.
18. method according to claim 10,11 or 12, it is characterised in that the mobile cloud computing halfpace is provided
Cloud computing management is carried out to external interface.
19. method according to claim 10,11 or 12, it is characterised in that when there is intelligent end in the distributed system
When the distributed system is exited at end, also include：
The mobile cloud computing halfpace is received and exits distributed system application from the Hadoop platform, described
Data resource migration and computing resource in the distributed system are carried out in Hadoop platform again to integrate；
After the intelligent terminal is exited successfully, the mobile cloud computing halfpace log.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611074053.3A CN106790403B (en) | 2016-11-29 | 2016-11-29 | Method for realizing mobile cloud computing intermediate platform and method for realizing distribution |
| PCT/CN2017/113647 WO2018099406A1 (en) | 2016-11-29 | 2017-11-29 | Method for realizing mobile cloud-computing intermediate platform and method for realizing distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611074053.3A CN106790403B (en) | 2016-11-29 | 2016-11-29 | Method for realizing mobile cloud computing intermediate platform and method for realizing distribution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106790403A true CN106790403A (en) | 2017-05-31 |
| CN106790403B CN106790403B (en) | 2022-01-25 |
Family
ID=58898589
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611074053.3A Active CN106790403B (en) | 2016-11-29 | 2016-11-29 | Method for realizing mobile cloud computing intermediate platform and method for realizing distribution |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106790403B (en) |
| WO (1) | WO2018099406A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018099406A1 (en) * | 2016-11-29 | 2018-06-07 | 中兴通讯股份有限公司 | Method for realizing mobile cloud-computing intermediate platform and method for realizing distribution |
| CN109246479A (en) * | 2018-10-09 | 2019-01-18 | 深圳市亿联智能有限公司 | A kind of cloud computing control mode based on Intelligent set top box |
| CN113254505A (en) * | 2021-06-17 | 2021-08-13 | 湖南视觉伟业智能科技有限公司 | Distributed data storage method, retrieval method, system and readable storage medium |
| CN114866565A (en) * | 2022-04-20 | 2022-08-05 | 北京红山信息科技研究院有限公司 | Software and hardware resource distribution system based on pass platform |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112905319B (en) * | 2021-02-04 | 2024-04-02 | 重庆惠科金渝光电科技有限公司 | Method, device and equipment for mobile cloud service position adjustment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110302302A1 (en) * | 2010-06-04 | 2011-12-08 | Electronics And Telecommunications Research Institute | Adaptive mobile cloud system using private virtual intance and construction method thereof |
| US20130198384A1 (en) * | 2012-01-27 | 2013-08-01 | MicroTechnologies LLC d/b/a Micro Tech | Transportable private cloud computing platform and associated method of use |
| US20130227558A1 (en) * | 2012-02-29 | 2013-08-29 | Vmware, Inc. | Provisioning of distributed computing clusters |
| CN103399787A (en) * | 2013-08-06 | 2013-11-20 | 北京华胜天成科技股份有限公司 | Map Reduce task streaming scheduling method and scheduling system based on Hadoop cloud computing platform |
| CN103414761A (en) * | 2013-07-23 | 2013-11-27 | 北京工业大学 | Mobile terminal cloud resource scheduling method based on Hadoop framework |
| CN103813213A (en) * | 2014-02-25 | 2014-05-21 | 南京工业大学 | Real-time video sharing platform and method based on mobile cloud computing |
| US20150026250A1 (en) * | 2013-10-08 | 2015-01-22 | Alef Mobitech Inc. | System and method of delivering data that provides service differentiation and monetization in mobile data networks |
| US20150134727A1 (en) * | 2013-11-12 | 2015-05-14 | Konkuk University Industrial Cooperation Corp. | Cloud-based data server providing home appliance management service and method thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105321124A (en) * | 2015-11-23 | 2016-02-10 | 南京信息工程大学 | Hadoop-based electric power cloud platform design scheme |
| CN106790403B (en) * | 2016-11-29 | 2022-01-25 | 中兴通讯股份有限公司 | Method for realizing mobile cloud computing intermediate platform and method for realizing distribution |
-
2016
- 2016-11-29 CN CN201611074053.3A patent/CN106790403B/en active Active
-
2017
- 2017-11-29 WO PCT/CN2017/113647 patent/WO2018099406A1/en active Application Filing
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110302302A1 (en) * | 2010-06-04 | 2011-12-08 | Electronics And Telecommunications Research Institute | Adaptive mobile cloud system using private virtual intance and construction method thereof |
| US20130198384A1 (en) * | 2012-01-27 | 2013-08-01 | MicroTechnologies LLC d/b/a Micro Tech | Transportable private cloud computing platform and associated method of use |
| US20130227558A1 (en) * | 2012-02-29 | 2013-08-29 | Vmware, Inc. | Provisioning of distributed computing clusters |
| CN103414761A (en) * | 2013-07-23 | 2013-11-27 | 北京工业大学 | Mobile terminal cloud resource scheduling method based on Hadoop framework |
| CN103399787A (en) * | 2013-08-06 | 2013-11-20 | 北京华胜天成科技股份有限公司 | Map Reduce task streaming scheduling method and scheduling system based on Hadoop cloud computing platform |
| US20150026250A1 (en) * | 2013-10-08 | 2015-01-22 | Alef Mobitech Inc. | System and method of delivering data that provides service differentiation and monetization in mobile data networks |
| US20150134727A1 (en) * | 2013-11-12 | 2015-05-14 | Konkuk University Industrial Cooperation Corp. | Cloud-based data server providing home appliance management service and method thereof |
| CN103813213A (en) * | 2014-02-25 | 2014-05-21 | 南京工业大学 | Real-time video sharing platform and method based on mobile cloud computing |
Non-Patent Citations (2)
| Title |
|---|
| 王小燕: "对Android的移动云计算技术的分析", 《自动化与仪器仪表》 * |
| 王慧娟，李淑凤: "浅谈Hadoop在移动云计算中的应用", 《北华航天工业学院学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018099406A1 (en) * | 2016-11-29 | 2018-06-07 | 中兴通讯股份有限公司 | Method for realizing mobile cloud-computing intermediate platform and method for realizing distribution |
| CN109246479A (en) * | 2018-10-09 | 2019-01-18 | 深圳市亿联智能有限公司 | A kind of cloud computing control mode based on Intelligent set top box |
| CN113254505A (en) * | 2021-06-17 | 2021-08-13 | 湖南视觉伟业智能科技有限公司 | Distributed data storage method, retrieval method, system and readable storage medium |
| CN113254505B (en) * | 2021-06-17 | 2021-10-08 | 湖南视觉伟业智能科技有限公司 | Distributed data storage method, retrieval method, system and readable storage medium |
| CN114866565A (en) * | 2022-04-20 | 2022-08-05 | 北京红山信息科技研究院有限公司 | Software and hardware resource distribution system based on pass platform |
| CN114866565B (en) * | 2022-04-20 | 2024-03-22 | 北京红山信息科技研究院有限公司 | Distribution system for software resources based on pass platform |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2018099406A1 (en) | 2018-06-07 |
| CN106790403B (en) | 2022-01-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106790403A (en) | Realize the method for mobile cloud computing halfpace and realize distributed method | |
| CN103152393B (en) | A kind of charging method of cloud computing and charge system | |
| CN103064742B (en) | A kind of automatic deployment system and method for hadoop cluster | |
| CN107566165B (en) | Method and system for discovering and deploying available resources of power cloud data center | |
| CN105323282A (en) | Enterprise application deployment and management system for multiple tenants | |
| CN106293820A (en) | Exploitation test O&M integral system, deployment, full dose and increment updating method | |
| CN105553741A (en) | Automatic deployment method for application system based on cloud computing | |
| CN104737135B (en) | The information processing terminal and synchronisation control means | |
| CN101593134A (en) | Virtual machine cpu resource distribution method and device | |
| CN104283960A (en) | System for achieving heterogeneous network storage virtualization integration and hierarchical management | |
| CN101420458B (en) | Multimedia content monitoring system, method and device based on content distributing network | |
| CN105282019A (en) | Service-based data gateway configurable method and system | |
| CN103530193A (en) | Method and device used for adjusting application process | |
| CN104660633A (en) | New media public service platform | |
| CN103200199A (en) | Out of band (OOB) data collection system | |
| CN103780685A (en) | Method for adaptive data sharing among diversified intelligent equipment | |
| WO2008091037A1 (en) | Realtime unification management information data conversion and monitoring apparatus and method for thereof | |
| CN103744618A (en) | Method and system for achieving team shared storage | |
| CN104219175B (en) | Data exchange and service calling system and method | |
| CN107196992A (en) | A kind of file data management system of law-enforcing recorder | |
| CN110011827A (en) | Towards doctor conjuncted multi-user's big data analysis service system and method | |
| CN104205730A (en) | Network element data access method, device and network management system | |
| CN101394397A (en) | Compression storage method capable of remote invoking used for mobile network, system thereof | |
| CN108696550B (en) | System and method for quickly building and replicating clouds | |
| CN102752374A (en) | System and method for storing and accessing power utilization efficacy data |
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 |