WO2018099406A1 - 实现移动云计算中间平台的方法及实现分布式的方法 - Google Patents

实现移动云计算中间平台的方法及实现分布式的方法 Download PDF

Info

Publication number
WO2018099406A1
WO2018099406A1 PCT/CN2017/113647 CN2017113647W WO2018099406A1 WO 2018099406 A1 WO2018099406 A1 WO 2018099406A1 CN 2017113647 W CN2017113647 W CN 2017113647W WO 2018099406 A1 WO2018099406 A1 WO 2018099406A1
Authority
WO
WIPO (PCT)
Prior art keywords
cloud computing
platform
mobile cloud
smart terminal
computing
Prior art date
Application number
PCT/CN2017/113647
Other languages
English (en)
French (fr)
Inventor
刘勇
陆小慧
张家明
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2018099406A1 publication Critical patent/WO2018099406A1/zh

Links

Images

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

Definitions

  • This article refers to, but is not limited to, big data technology, especially a method for implementing mobile cloud computing intermediate platform and a distributed method.
  • Hadoop is a distributed system infrastructure developed by the Apache Foundation. Users can develop distributed programs without knowing the underlying details of the distribution. Take full advantage of the power of the cluster for high-speed computing and storage.
  • Hadoop Distributed File System (HDFS) is highly fault-tolerant and designed for deployment on lower-cost PC servers than minicomputers; HDFS provides high throughput (high throughput) to access application data, suitable for applications with large data sets. HDFS relaxes the requirements of POSIX and can stream access data in the file system.
  • the core design of the Hadoop framework includes: HDFS and MapReduce.
  • HDFS provides storage for massive amounts of data
  • MapReduce provides calculations for massive amounts of data.
  • Hadoop is open source free, hardware is cheap, and development is convenient. Therefore, it dominates the big data market, making Hadoop's market share increase year by year.
  • the embodiments of the present invention provide a method for implementing a mobile cloud computing intermediate platform and a distributed implementation method, which can enable the Android system to support the operation of Hadoop.
  • An embodiment of the present invention provides a method for implementing a mobile cloud computing intermediate platform, including:
  • Cloud computing management based on network configuration information and integrated computing and storage resources
  • the mobile cloud computing intermediate platform is established between the Hadoop platform and the Android Android system.
  • the method further includes: providing the Hadoop platform with a Java library that is missing from the Android system.
  • the network configuration includes:
  • a smart address is allocated to the smart terminal that installs the mobile cloud computing intermediate platform
  • the obtaining the IP address according to the connection manner of the distributed system to which the smart terminal belongs includes:
  • connection mode is a wired connection, searching for an IP address of all intelligent terminals of the distributed system, and processing the largest IP address as the IP address;
  • the IP address is automatically allocated by the wireless network dynamic host configuration protocol DHCP;
  • connection mode is a custom rack mode
  • the IP address is obtained according to the slot position.
  • the managing and storing the resources, and performing resource integration processing based on the computing and storage resources includes:
  • Retrieving the task allocation method of the Hadoop platform according to the confirmed resource usage status when the storage size exceeds the upper limit of the set limit, the non-receiving operation is performed, and when the computing power exceeds the upper limit of the set limit, the computing task is not accepted; when the storage size is low When setting the lower limit of the limit, write priority until it reaches Setting an upper limit of the limit; when the computing power is lower than the lower limit of the set limit, the computing task is preferentially accepted until the upper limit of the set limit is reached; the intelligent terminal in the distributed system is between the upper limit and the lower limit Random assignment when getting a task.
  • the method further includes: providing hardware health monitoring to the smart terminal in the distributed system for installing the mobile cloud computing intermediate platform, and providing a repair operation.
  • the method further includes: performing a reconnection or restart operation, and prompting to replace the hardware after the preset number of reconnections or restarts still fails.
  • the method further includes: performing a first-in first-out FIFO storage when the system log is greater than the set range value.
  • the embodiment of the invention further provides an apparatus for implementing a mobile cloud computing intermediate platform, comprising at least a memory and a processor, wherein the memory stores the following executable instructions:
  • the mobile cloud computing intermediate platform is established between the Hadoop platform and the Android Android system.
  • the embodiment of the invention further provides a method for implementing distributed, comprising:
  • the mobile cloud computing intermediate platform performs data resource migration and computing resource integration in the distributed system to which the smart terminal belongs in the Hadoop platform;
  • the mobile cloud computing intermediate platform receives storage and computing resources of the smart terminals in the distributed system fed back from the Hadoop platform, and performs cloud computing management.
  • the method further includes:
  • the mobile cloud computing intermediate platform re-implements data resource migration and computing resource integration according to the storage and computing resources of the newly accessed smart terminal.
  • the method further includes:
  • a new smart terminal is connected to the network where the distributed system is located, and the mobile cloud computing intermediate platform and the Hadoop platform are acquired and installed;
  • the new smart terminal uses the installed mobile cloud computing intermediate platform to configure the Hadoop system network according to the connection mode of the network where the distributed system is located.
  • the method further includes obtaining an IP address according to a connection manner of the distributed system to which the smart terminal belongs, including:
  • connection mode is a wired connection, searching for an IP address of all intelligent terminals of the distributed system, and processing the largest IP address as the IP address;
  • the IP address is automatically allocated by the wireless network dynamic host configuration protocol DHCP;
  • connection mode is a custom rack mode
  • the IP address is obtained according to the slot position.
  • the data resource migration and the computing resource integration in the distributed system further include:
  • the mobile cloud computing intermediate platform provides a Java library that is missing from the Android system of the smart terminal, and performs hardware selection and loading on the Android system of the smart terminal;
  • the mobile cloud computing intermediate platform allocates an IP address to the smart terminal according to the distributed system connection manner returned by the smart terminal; and writes the configuration file of the Hadoop platform.
  • the performing data resource migration and computing resource integration includes:
  • a smart terminal whose storage size exceeds the upper limit of the set limit does not accept the write operation, and the functional terminal whose computing power exceeds the upper limit of the set limit does not accept the calculation task;
  • the write is prioritized until the upper limit of the set limit is reached; when the computing power of the intelligent terminal is lower than the lower limit of the set limit, the calculation task is preferentially accepted until the set limit is reached.
  • the storage size and the computing power are randomly assigned when the smart terminal acquires a task between the upper limit and the lower limit.
  • the method further includes: the mobile cloud computing intermediate platform uses a FIFO storage manner to store logs.
  • the mobile cloud computing intermediate platform performs hardware health monitoring according to a heartbeat situation of the smart terminal accessed by the distributed system fed back by the Hadoop platform.
  • the mobile cloud computing intermediate platform provides an external interface for cloud computing management.
  • the method further includes:
  • the mobile cloud computing intermediate platform receives an exit distributed system application from the Hadoop platform, and performs data resource migration and computing resource integration in the distributed system again in the Hadoop platform;
  • the mobile cloud computing intermediate platform After the smart terminal exits successfully, the mobile cloud computing intermediate platform records the log.
  • the mobile cloud computing intermediate platform is established between the Hadoop and the Android system, and is used for ensuring the adaptation and operation between the Hadoop and the Android system, so that the Hadoop implements the normal operation on the Android system, that is, the Android system supports. Hadoop runs.
  • the intelligent terminal realizes a very convenient distributed system for building, joining, and removing a wireless network or a wired network, and particularly realizes that the use is cheaper and more convenient to replace. Intelligent terminals form a distributed cluster.
  • FIG. 1 is a schematic structural diagram of a mobile cloud computing intermediate platform according to an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of a working principle of a mobile cloud computing intermediate platform according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of implementing data storage in an HDFS cluster based on an Android system according to an embodiment of the present invention
  • FIG. 4 is a schematic flowchart of implementing an HDFS cluster based on an Android system according to an embodiment of the present invention
  • FIG. 5 is a schematic structural diagram of a computing module in an intelligent terminal after being modified according to an embodiment of the present invention
  • FIG. 6 is a schematic structural diagram of a storage module in an intelligent terminal after being modified according to an embodiment of the present invention
  • FIG. 7 is a schematic diagram of networking of a smart terminal adopting a WIFI connection mode according to an embodiment of the present invention.
  • FIG. 8 is a schematic flowchart diagram of an embodiment of managing an intelligent terminal in a cluster based on the connection manner shown in FIG. 7 according to an embodiment of the present invention
  • FIG. 9 is a schematic diagram of networking of a smart terminal using a mobile network connection manner according to an embodiment of the present invention.
  • FIG. 10 is a schematic flowchart diagram of an embodiment of managing an intelligent terminal in a cluster based on the connection manner shown in FIG. 9 according to an embodiment of the present invention
  • FIG. 11 is a schematic diagram of a networking in which a smart terminal adopts a wired connection manner according to an embodiment of the present invention
  • FIG. 12 is a schematic flowchart diagram of an embodiment of managing an intelligent terminal in a cluster based on the connection manner shown in FIG. 11 according to an embodiment of the present invention
  • FIG. 13 is a top plan view of a rack in a customized rack structure according to an embodiment of the present invention.
  • FIG. 14 is a side view of a rack in a customized rack structure according to an embodiment of the present invention.
  • 15 is a schematic diagram of implementing hardware replacement in a customized rack structure according to an embodiment of the present invention.
  • 16 is a schematic flowchart of an embodiment of a mobile cloud computing intermediate platform application according to an embodiment of the present invention.
  • FIG. 17 is a flowchart of a method for implementing a mobile cloud computing intermediate platform according to an embodiment of the present invention.
  • FIG. 18 is a flowchart of a method for implementing distributed implementation of a mobile cloud computing intermediate platform according to an embodiment of the present invention.
  • CPU Central Processing Unit
  • the CPU processor used in the smart terminal is inexpensive.
  • the price difference between the two is about 30 times, and the power consumption is about 1000 times different.
  • the performance gap between the two is on average to a single core, and the central processor is only about 5 times that of a mobile terminal (such as a mobile phone) processor.
  • a cluster of 100 intelligent terminals can cost between 1 and 20,000 yuan, and the physical volume is equivalent to that of a normal PC server.
  • the computing performance is 5-10 times that of a PC server. Can be controlled from 1/10 to 1/20 of the server.
  • the mobile cloud computing intermediate platform is an intermediate layer between a Hadoop platform and an Android system. Adaptation and operation between the Hadoop platform and the Android system.
  • the method includes at least: a hardware driver module, a Java library support module, a configuration module, a computing storage resource management module, and a cloud computing task scheduling module; wherein
  • the hardware driver module is set to load the hardware driver for the operating system.
  • the driver is mainly used for network controller services.
  • the network controller is a USB to RJ45 signal chip; when the communication is optical fiber communication, the network controller is a photoelectric conversion chip.
  • the Java library support module is set to provide a Java library that is missing from the Android system for the Hadoop platform.
  • the Android system itself supports a part of Java functions, and the Java library support module here is used to complement the missing Java libraries in the Android system.
  • the configuration module is configured to perform network configuration; the operating configuration of the Hadoop platform is automatically set according to the computing capacity and storage size of the intelligent terminal that is added to the distributed system.
  • the configuration module is configured to implement network configuration as follows:
  • the mobile cloud computing intermediate platform where the configuration module itself is located is assigned a fixed address for the smart terminal installing the mobile cloud computing intermediate platform when the first startup is started (all IP addresses are initial when all machines are started);
  • connection mode of the distributed system to which the smart terminal belongs such as wired, wireless, or custom rack type
  • IP address when it is a wired connection, search for the IP addresses of all machines of the current network, and then The largest IP address is processed: +1 in the fourth segment of the largest IP address, when added to 253, it is incremented in the third segment of the IP address, and the fourth segment starts counting from 1; when it is a wireless connection, IP
  • the address is automatically assigned by the Dynamic Host Configuration Protocol (DHCP). It does not need to consider the IP address allocation. It only needs to read the assigned IP address and write it into the Hadoop configuration file. The IP address of the device in the same slot is unchangeable.
  • DHCP Dynamic Host Configuration Protocol
  • the computing resource management module is configured to be integrated into the mobile cloud computing intermediate platform where the computing storage resource management module is located on the basis of the existing Hadoop computing and storage control, and manage the calculated and stored resources, based on the computing and storage resources.
  • the implementation includes: when the distributed system receives the external write and calculation tasks, first confirms the resource usage status of the machine in the current distributed system, and then retrieves the task allocation method of the Hadoop platform: the storage that has exceeded the upper limit of the set limit will not be accepted.
  • the calculation task will not be accepted; when the storage is lower than the lower limit of the set limit, the priority will be written until the upper limit of the set limit is reached; when the calculation is lower than the lower limit of the set limit, the priority will be given.
  • the calculation task is accepted until the upper limit of the set limit is reached; other machines between the upper and lower limits are randomly assigned when acquiring the task.
  • the cloud computing task scheduling module is configured to provide to the external interface, and perform cloud computing management according to the network configuration information and the integrated computing and storage resources, and the management includes starting, stopping, pausing, timing, and the like.
  • the mobile cloud computing intermediate platform of the embodiment of the present invention further includes:
  • the health monitoring module is configured to provide hardware health monitoring for the smart terminal and provide repair operations. It should be noted that the health monitoring module only monitors the hardware and death. When the hardware has a problem, it will reconnect or restart. After multiple reconnections or restart failures, the system prompts to replace the hardware.
  • the monitoring function can use the heartbeat detection mechanism to monitor hardware life and death, and use temperature monitoring to run hardware health alarms.
  • the mobile cloud computing intermediate platform of the embodiment of the present invention further includes:
  • the log processing module is configured to store the log in a first in first out (FIFO) manner when the system log is greater than the set range value, that is, to delete the forward log and then store the latest log.
  • FIFO first in first out
  • FIG. 1 only shows the basic Hadoop component of the mobile cloud computing intermediate platform according to the embodiment of the present invention.
  • Other components can be installed by themselves, whether other components are included, and how much is included, and are not used to limit the scope of protection of the present invention. .
  • the embodiment of the present invention further provides a method for implementing a mobile cloud computing intermediate platform.
  • the method includes: setting a mobile cloud computing intermediate platform between a Hadoop platform and an Android Android system, and further comprising:
  • S1701 performs network configuration; sets Hadoop flat according to smart terminals added to the distributed system. Operational configuration of the station;
  • S1702 manages the calculated and stored resources, and performs resource integration processing based on the computing and storage resources;
  • the S1703 performs cloud computing management based on network configuration information and integrated computing and storage resources.
  • the method of the embodiment of the present invention further includes: providing a Java library missing from the Android system for the Hadoop platform.
  • the network configuration performed includes:
  • the smart terminal for installing the mobile cloud computing intermediate platform is assigned a fixed address; and the IP address is obtained according to the connection mode of the distributed system to which the smart terminal belongs. among them,
  • Obtaining an IP address according to the connection mode of the distributed system to which the smart terminal belongs includes:
  • connection mode is a wired connection, searching for the IP addresses of all intelligent terminals of the distributed system, and processing the largest IP address as the IP address;
  • the IP address is automatically assigned by the wireless network DHCP;
  • connection mode is the custom rack mode
  • IP address is obtained according to the slot.
  • managing the calculated and stored resources, and performing resource integration processing based on the computing and storage resources includes:
  • the task allocation method of the Hadoop platform is retrieved according to the confirmed resource usage status: when the storage size exceeds the upper limit of the set limit, the write operation is not accepted, and when the calculation capability exceeds the upper limit of the set limit, the calculation task is not accepted; when the storage size is lower than the set When the lower limit is set, the priority is written until the upper limit of the set limit is reached; when the computing power is lower than the lower limit of the set limit, the calculation task is preferentially accepted until the upper limit of the set limit is reached; the storage size and the storage system A smart terminal with a computing power between the upper and lower limits performs random allocation when acquiring a task.
  • the method for implementing the mobile cloud computing intermediate platform in the embodiment of the present invention further includes:
  • reconnecting or restarting prompting to replace the hardware after the preset number of reconnections or restarts still fails.
  • the method for implementing the mobile cloud computing intermediate platform in the embodiment of the present invention further includes: performing FIFO storage when the system log is greater than the set range value.
  • An embodiment of the present invention further provides an apparatus for implementing a mobile cloud computing intermediate platform, including at least a memory and a processor, where
  • the following executable instructions are stored in the memory: the mobile cloud computing intermediate platform is established between the Hadoop platform and the Android Android system; the network configuration is performed; the operational configuration of the Hadoop platform is set according to the intelligent terminal added to the distributed system; and the calculation and storage are performed.
  • the mobile cloud computing intermediate platform established by using the embodiment of the present invention to implement the distributed method includes:
  • S1801 acquires and installs a mobile cloud computing intermediate platform and a Hadoop platform to the smart terminal;
  • the S1802 mobile cloud computing intermediate platform performs data resource migration and computing resource integration in the distributed system to which the smart terminal belongs in the Hadoop platform;
  • the S1803 mobile cloud computing intermediate platform receives storage and computing resources of the smart terminal in the distributed system fed back from the Hadoop platform, and performs cloud computing management.
  • FIG. 1 of the embodiment of the present invention The working principle of each module after installing the mobile cloud computing intermediate platform shown in FIG. 1 of the embodiment of the present invention is taken as an example to describe the working principle of the mobile cloud computing intermediate platform in the embodiment of the present invention.
  • 2 is a schematic flowchart of a working principle of a mobile cloud computing intermediate platform according to an embodiment of the present invention, as shown in FIG. 2, including:
  • the smart terminal acquires and installs the mobile cloud computing intermediate platform and the Hadoop platform according to the embodiment of the present invention
  • the Java library support module of the mobile cloud computing intermediate platform provides the Java library that is lacking in the Android system of the smart terminal
  • the hardware driver module of the cloud computing intermediate platform performs hardware selection and loading on the Android system of the smart terminal; the implementation of the present invention
  • the configuration module of the mobile cloud computing intermediate platform allocates an IP address to the smart terminal: the smart terminal returns the distributed system connection type to the configuration module of the mobile cloud computing intermediate platform according to the embodiment of the present invention;
  • the configuration module performs configuration of the corresponding IP address according to different network connection modes of the intelligent terminal: when it is a wired connection, searches for the IP address of all machines on the current network, and then processes the largest IP address: in the fourth segment of the largest IP address.
  • the fourth segment starts counting from 1; when it is wireless connection, the IP address is automatically assigned by the wireless network DHCP; when it is customized rack mode, Obtain an IP address configuration based on the slot.
  • the intelligent terminal reads the assigned IP address and writes the configuration file of the Hadoop platform and takes effect;
  • the computing storage resource management module of the mobile cloud computing intermediate platform in the embodiment of the present invention performs data resource migration and computing resource integration in the distributed system in the Hadoop platform, and the Hadoop platform feeds back the storage and computing resources of the intelligent terminal to the implementation of the present invention.
  • the computing storage resource management module of the mobile cloud computing intermediate platform performs data resource migration and computing resource integration according to the storage and computing resources of the newly accessed smart terminal: the storage has exceeded the upper limit of the set limit. Accept the write operation, the calculation will not accept the calculation task beyond the upper limit of the set limit; when the storage is lower than the lower limit of the set limit, the priority will be written until the upper limit of the set limit is reached; when the calculation is lower than the lower limit of the set limit The computing task is preferentially accepted until the upper limit of the set limit is reached; other machines between the upper and lower limits are randomly assigned when acquiring the task;
  • the log processing module of the mobile cloud computing intermediate platform performs FIFO storage according to the embodiment of the present invention; the health monitoring module of the mobile cloud computing intermediate platform performs hardware health monitoring and the feedback of the Hadoop platform feedback access.
  • the heartbeat situation of the terminal is used to detect the hardware and death, and the node temperature of the intelligent terminal is fed back to monitor the running health condition.
  • the cloud computing task scheduling module of the mobile cloud computing intermediate platform is provided to the external interface for cloud computing management.
  • the Hadoop platform applies to the computing storage resource management module of the mobile cloud computing intermediate platform of the embodiment of the present invention to apply for exiting the distributed system, and the computing storage resource management of the mobile cloud computing intermediate platform is implemented in the embodiment of the present invention.
  • Module will be on the Hadoop platform
  • the data resource migration and the computing resource integration in the distributed system are performed again, and after the exit is successful, the log is recorded in the log processing module of the mobile cloud computing intermediate platform in the embodiment of the present invention.
  • the Hadoop platform based on the Android system in the embodiment of the present invention is the mobile cloud computing intermediate platform shown in FIG. 1 , on the one hand, the Hadoop platform realizes the normal operation on the Android system, and the Android system supports the Hadoop. The operation of the platform.
  • health monitoring of the Hadoop platform runtime software and hardware is provided to ensure that the repair of the software or the replacement of the hardware is triggered in the event of corruption.
  • the intelligent terminal realizes a distributed system that is very convenient to set up, join, and remove a wireless network or a wired network. Thereby, a distributed system that uses a smart terminal that is cheaper and more convenient to replace is realized.
  • the embodiment of the present invention is described by using an intelligent terminal as a mobile phone terminal.
  • the smart terminal of the embodiment of the present invention includes: a mobile phone terminal running the Android system, a PAD, a set top box, and each type of wired and wireless handheld terminal running the Android system.
  • the Hadoop platform uses the mobile cloud computing intermediate platform to invoke one or more application interface (API) functions on the underlying Android system to perform data reading and writing, calculation, and the like.
  • API application interface
  • the package of the Android API interface includes:
  • For intelligent terminal memory data read and write it can include:
  • getFileDir() used to obtain the location information of the Hadoop platform to store data in the smart terminal memory, such as: /data/data/Hadoop/files;
  • getCacheDir() used to obtain the location information of the Hadoop platform in the smart terminal memory cache data, such as: /data/data/Hadoop/cache;
  • openFileOutput (String name, int mode): used to directly obtain the output stream of the /data/data/Hadoop/files/name file, where mode is the permission when writing the file;
  • Context.MODE_PRIVATE is the private mode (or called the default mode), can only be accessed by the application itself and the smart terminal of the same group; the written content covers the original file content;
  • Context.MODE_APPEND is an append mode (or private mode), which can only be accessed by the application itself and the smart terminal of the same group; if the file exists, the content is appended, if the file does not exist, the file is newly created and the content is written;
  • getExternalStorageDirectory() used to obtain the SDcard location information of the smart terminal where the Hadoop platform is located, such as /storage/sdcard;
  • getExternalStorageState() used to obtain the current state of the SDcard of the smart terminal where the Hadoop platform is located. The more commonly used one is MEDIA_MOUNTED;
  • FileInputStream() used to read the file
  • the Hadoop calculation uses the addition, subtraction, multiplication, and division of the Java library.
  • the HDFS cluster based on the Android system has two types of nodes and operates in a "manager-worker" mode: a master node (NameNode) as a manager and a plurality of slaves as workers ( Slave) Node (DataNode). Each NameNode and DataNode corresponds to one intelligent terminal.
  • the NameNode is mainly responsible for managing the HDFS file system, including namespace management and block management.
  • the DataNode is mainly configured to store data files.
  • HDFS will split a file into blocks, which may be stored in a DataNode.
  • the DataNode is responsible for reading and writing the actual underlying files. If the client program initiates a command to read files on the HDFS, the files are first divided into blocks, and the NameNode will tell the Client that the block data is stored. Which DataNodes are on, so that the Client can directly interact with the DataNode.
  • FIG. 3 is a schematic flowchart of implementing data storage in an HDFS cluster based on an Android system according to an embodiment of the present invention.
  • the file is uploaded to a distributed cluster for storage.
  • the NameNode is responsible for managing metadata of all files stored on the HDFS, and confirming the client. Request, and record the name of the file and the collection of data nodes that store the file, the DataNode. The information of these records is stored in a file allocation table in memory.
  • the client sends a request to the NameNode, indicating that the "swim.txt" file is to be written to the HDFS, as shown in FIG. 3, including:
  • Step 1 The client sends a message to the NameNode indicating that the "swim.txt" file is to be written.
  • Step 2 The NameNode returns a message to the client, informing the client to write the "swim.txt" file to DataNode A, DataNode B and DataNode D, and directly contact DataNode B;
  • Step 3 The client sends a message to the DataNode B, instructing the DataNode B to save a "swim.txt" file, and send a copy to DataNode A and DataNode D respectively;
  • Step 4 The DataNode B sends a message to the DataNode A, instructing the DataNode A to save a "swim.txt" file and send a copy to the DataNode D;
  • Step 5 DataNode A sends a message to DataNode D, instructing DataNode D to save a "swim.txt" file;
  • Step 6 The DataNode D sends a confirmation message to the DataNode A;
  • Step 7 DataNode A sends a confirmation message to DataNode B;
  • Step 8 The DataNode B sends a confirmation message to the client, indicating that the writing of the "swim.txt" file is completed.
  • a "swim.txt" file is stored on the data nodes A, DataNode B, and DataNode D of the distributed cluster (ie, smart terminals).
  • each MapReduce task is initialized to a Job, and each Job can be further divided into Two phases: the Map phase and the Reduce phase. These two phases are represented by two functions, the Map function and the Reduce function.
  • the Map function receives an input of the form ⁇ key, value> and then produces an intermediate output of the form ⁇ key, value>;
  • the Reduce function receives an input of the form ⁇ key, (list of values)>, and then The value collection is processed.
  • Each Reduce generates 0 or 1 output, and the output of Reduce is also in the form of ⁇ key, value>.
  • a common smart terminal can be configured after the Hadoop platform is configured by using the mobile cloud computing intermediate platform of the embodiment of the present invention after the mobile cloud computing intermediate platform and the Hadoop platform of the embodiment of the present invention are installed.
  • Use of one or more components on top of Hadoop For example, in the office, you can centralize your smart terminals into a cluster to share computing resources and storage resources. In this way, the amount of calculation that one's smart terminal can't do can be set up by a smart terminal of a dozen people, and the calculation amount that cannot be completed can be completed on the mobile cloud computing intermediate platform paper of the embodiment of the present invention.
  • the wireless connection includes: WIFI network, mobile network (such as 2G, 3G, 4G, etc.) two cluster formation methods.
  • the cluster hardware source using the wired connection method can be a customized intelligent terminal, or can be modified by using an existing smart terminal for wired communication.
  • the transformation of the existing intelligent terminal mainly includes two parts: on the one hand, the USB interface is modified, and the USB type communication is converted into a twisted pair or optical fiber line communication; on the other hand, the storage module includes at least a ROM, an SD card, and interface. Storage module damage is replaceable.
  • FIG. 5 is a schematic structural diagram of a computing module in an intelligent terminal according to an embodiment of the present invention. As shown in FIG. 5, the method includes at least a central processing unit (CPU) chipset and a random access memory (RAM, Random- Access Memory), network controller. Externally includes three interfaces: power supply, network port, and storage interface.
  • CPU central processing unit
  • RAM Random- Access Memory
  • the USB interface may include: USB2.0 or USB3.0, which are respectively used for the twisted pair and the optical fiber. Therefore, there are two different types of communication schemes for USB.
  • the network controller adopts the common USB to network card scheme, and if it is a fiber, the photoelectric conversion chip can be added.
  • the network controller is a signal conversion chip, such as: USB2.0 to RJ45 signal chip such as AX88772B; USB 3.0-to-Gigabit Ethernet control chip such as RTL8153.
  • FIG. 6 is a schematic structural diagram of a storage module in an intelligent terminal according to an embodiment of the present invention. As shown in FIG. 6 , at least two built-in components: a ROM and an SD card are externally displayed as a storage interface. The module is damaged and can be replaced.
  • Hadoop is a versatile big data platform.
  • the Hadoop-based implementation can use the technical solution provided by the embodiment of the present invention, and the PC server is transferred to the smart terminal installed with the Android system.
  • the following describes an embodiment of the management of the intelligent terminal in the cluster according to the different network connection manners of the smart terminal, in conjunction with the mobile cloud computing intermediate platform according to the embodiment of the present invention.
  • FIG. 7 is a schematic diagram of a networking diagram of a smart terminal adopting a WIFI connection mode according to an embodiment of the present invention
  • FIG. 8 is a schematic flowchart diagram of an embodiment of managing an intelligent terminal in a cluster based on the connection manner shown in FIG. 7 and FIG. 8 , in this embodiment, taking a new node as an intelligent terminal as an example, the steps of adding the smart terminal to the cluster include:
  • Step 800 The smart terminal of the WIFI network to be added to the existing cluster is successfully connected to the WIFI network.
  • Step 801 The smart terminal of the WIFI network to join the existing cluster downloads and installs the mobile cloud computing intermediate platform and the Hadoop platform through the Android system.
  • Step 802 Configure a Hadoop platform system network by using a mobile cloud computing intermediate platform.
  • the implementation process is shown in Figure 2 and will not be described here.
  • connection mode (Connection) parameter 1
  • Connection parameter is 2
  • the connection parameter is 3
  • the Connection parameter is 3, indicating that it is wired connection.
  • the mobile cloud computing intermediate platform allocates an IP address to the newly added node in the embodiment (such as the smart terminal in step 800) according to the DHCP mechanism in the WIFI network; The IP address is written to the cluster configuration file.
  • Step 803 The primary node in the existing cluster starts the Hadoop cluster component and starts successfully.
  • Step 804 The primary node in the existing cluster performs data resource migration and computing resource balancing in the cluster.
  • the mobile cloud computing intermediate platform in the primary node performs data resource migration and computing resource integration according to the storage and computing resources of the new node (such as the smart terminal). For example, assume that the storage resource limit of each node is 80% of the total storage resources and the storage resource lower limit is 5% of the total storage resources; assume that the CPU computing resource limit of each node is 80% of CPU utilization and CPU calculation The lower limit of the resource is 5% of the CPU utilization; among them, the upper and lower limits can be adjusted according to actual needs.
  • the mobile cloud computing intermediate platform performs the following operations according to the set upper and lower limits:
  • the write will be prioritized until the upper limit of the storage resource is reached;
  • the computing task When the calculation of the newly added node is lower than the lower limit of the CPU computing resource, the computing task will be preferentially accepted until the CPU computing resource upper limit is reached;
  • Others are between the upper and lower limits and can be randomly assigned when acquiring a task.
  • Step 805 The master node in the existing cluster invokes the mobile cloud computing intermediate platform interface, and submits a Hadoop task to run.
  • Step 806 The data node in the cluster completes the task calculation, and feeds the task result to the master node.
  • a certain data node such as the smart terminal joining the existing cluster in step 800, applies to exit the cluster, including the following steps:
  • Step 807 The intelligent terminal applies to the primary node in the cluster to exit the cluster.
  • Step 808 The primary node performs data resource migration and computing resource integration. The implementation is as described in step 804, and details are not described herein again.
  • Step 809 After the resource integration is completed, the smart terminal is allowed to exit the cluster.
  • FIG. 9 is a schematic diagram of a networking diagram of a smart terminal using a mobile network connection manner according to an embodiment of the present invention
  • FIG. 10 is a schematic flowchart of an embodiment of managing an intelligent terminal in a cluster based on the connection manner shown in FIG. 9 according to an embodiment of the present invention
  • the new node is an intelligent terminal as an example.
  • the steps of adding the smart terminal to the cluster include:
  • Step 1000 The mobile terminal that wants to join the existing cluster, such as 2G, 3G, 4G, etc., successfully connects to the mobile network.
  • existing cluster such as 2G, 3G, 4G, etc.
  • Step 1001 The smart terminal of the mobile network to join the existing cluster downloads and installs the mobile cloud computing intermediate platform and the Hadoop platform through the Android system.
  • Step 1002 Configure the Hadoop system network using the mobile cloud computing intermediate platform.
  • the implementation process is shown in Figure 2 and will not be described here.
  • connection mode (Connection) parameter 1
  • Connection parameter is 2
  • the connection parameter is 3
  • the Connection parameter is 3, indicating that it is wired connection.
  • the mobile cloud computing intermediate platform allocates an IP address to the newly added node in the embodiment (such as the smart terminal in step 1000) according to the DHCP mechanism in the mobile network; The IP address is written to the cluster configuration file.
  • Step 1003 The primary node in the existing cluster starts the Hadoop cluster component and starts successfully.
  • Step 1004 The primary node in the existing cluster performs data resource migration and computing resource balancing in the cluster.
  • the mobile cloud computing intermediate platform in the primary node performs data resource migration and computing resource integration according to the storage and computing resources of the new node (such as the smart terminal).
  • the upper limit of the storage resource of each node is 80% of the total storage resources and the lower limit of the storage resources is 5% of the total storage resources.
  • the CPU computing resource limit of each node is the CPU utilization.
  • the 80% and CPU computing resource lower limit is 5% of the CPU utilization; among them, the upper and lower limits can be adjusted according to actual needs.
  • the mobile cloud computing intermediate platform performs the following operations according to the set upper and lower limits:
  • the write will be prioritized until the upper limit of the storage resource is reached;
  • the computing task When the calculation of the newly added node is lower than the lower limit of the CPU computing resource, the computing task will be preferentially accepted until the CPU computing resource upper limit is reached;
  • Others are between the upper and lower limits and can be randomly assigned when acquiring a task.
  • a data compression (Compress) parameter may be set in consideration of a transmission rate and a traffic tariff.
  • Compress parameter When the Compress parameter is 1, it indicates that compression is required.
  • Compress parameter When the Compress parameter is 0, it indicates that it is unnecessary. Compress.
  • the Compress parameter is automatically set to 1, indicating that compression is required during data transmission to obtain a high transmission rate, thereby reducing the traffic charge;
  • the Compress parameter is automatically set to 0, and the data transfer does not need to be compressed to reduce the CPU operation pressure.
  • Step 1005 The master node in the existing cluster invokes the mobile cloud computing intermediate platform interface, and submits a Hadoop task to run.
  • Step 1006 The data node in the cluster completes the task calculation, and feeds the task result to the master node.
  • a certain data node such as the smart terminal joining the existing cluster in step 1000, requests to exit the cluster, including the following steps:
  • Step 1007 The intelligent terminal applies to the primary node in the cluster to exit the cluster.
  • Step 1008 The primary node performs data resource migration and computing resource integration. The implementation is as described in step 1004, and details are not described herein again.
  • Step 1009 After the resource integration is completed, the smart terminal is allowed to exit the cluster.
  • FIG. 11 is a schematic diagram of a networking diagram of a smart terminal in a wired connection manner according to an embodiment of the present invention.
  • FIG. 12 is a schematic flowchart diagram of an embodiment of managing an intelligent terminal in a cluster according to the connection mode shown in FIG. 11 and FIG. 12, in this embodiment, taking a new node as an intelligent terminal as an example, the steps of adding the smart terminal to the cluster include:
  • Step 1200 The smart terminal to be added to the existing cluster successfully connects to the switch in the cluster by using a twisted pair cable or a fiber.
  • the smart terminal to be added to the existing cluster uses the USB 2.0 interface. Therefore, the switch of the cluster is connected by using a twisted pair cable.
  • Step 1201 The smart terminal of the mobile network to join the existing cluster downloads and installs the mobile cloud computing intermediate platform and the Hadoop platform through the Android system.
  • Step 1202 Configure the Hadoop system network using the mobile cloud computing intermediate platform.
  • the implementation process is shown in Figure 2 and will not be described here.
  • connection mode (Connection) parameter 1
  • Connection parameter is 2
  • the connection parameter is 3
  • the Connection parameter is 3, indicating that it is wired connection.
  • the mobile cloud computing intermediate platform searches for the IP addresses of all nodes in the current cluster network, and then processes the largest IP address: +1 in the fourth segment of the largest IP address, when added At 253, the third segment of the IP address is incremented by one, and the fourth segment is counted from 1; the assigned IP address is written into the cluster configuration file.
  • Step 1203 The primary node in the existing cluster starts the Hadoop cluster component and starts successfully.
  • Step 1204 The primary node in the existing cluster performs data resource migration and computing resource balancing in the cluster.
  • the mobile cloud computing intermediate platform in the primary node performs data resource migration and computing resource integration according to the storage and computing resources of the new node (such as the smart terminal).
  • the upper limit of the storage resource of each node is 80% of the total storage resources and the lower limit of the storage resources is 5% of the total storage resources.
  • the CPU computing resource limit of each node is the CPU utilization.
  • the 80% and CPU computing resource lower limit is 5% of the CPU utilization; among them, the upper and lower limits can be adjusted according to actual needs.
  • the mobile cloud computing intermediate platform performs the following operations according to the set upper and lower limits:
  • the write will be prioritized until the upper limit of the storage resource is reached;
  • the computing task When the calculation of the newly added node is lower than the lower limit of the CPU computing resource, the computing task will be preferentially accepted until the CPU computing resource upper limit is reached;
  • Others are between the upper and lower limits and can be randomly assigned when acquiring a task.
  • Step 1205 The master node in the existing cluster invokes the mobile cloud computing intermediate platform interface, and submits a Hadoop task to run.
  • Step 1206 The data node in the cluster completes the task calculation, and feeds the task result back to the master node.
  • step 1200 joins an intelligent terminal of an existing cluster to apply for exiting the cluster, including the following steps:
  • Step 1207 The intelligent terminal applies to the primary node in the cluster to exit the cluster.
  • Step 1208 The primary node performs data resource migration and computing resource integration. The implementation is as described in step 1004, and details are not described herein again.
  • Step 1209 After the resource integration is completed, the smart terminal is allowed to exit the cluster.
  • This embodiment is a case where the cluster model is determined to be a customized rack structure after the network is connected.
  • the rack converts the AC power into DC power to supply power to all the devices.
  • the entire system shown in Figure 1 is packaged as an image and stored directly in the ROM.
  • the installed hardware of the system is stored in the rack entrance in the rack where the backup is located. Initial installation because of The number of devices to be installed is large, and manual hardware supply is required. Subsequent backup devices preferably store 5-10 devices automatically.
  • FIG. 13 is a top view of the rack in the customized rack structure according to the embodiment of the present invention.
  • the low-speed rotation of the ring brings the effect of the air-cooling heat dissipation of the conventional equipment.
  • the low-speed rotation of the ring realizes the docking of the calculation storage backup module channel, so that the rack will be The replacement module is shipped in place; at the same time, centrifugal force can automatically remove the damaged hardware.
  • the hardware connectivity Rack Switching device automatically obtains network parameters and joins the distributed cluster.
  • the device that first joins the cluster is the master node, and the distributed computing task can be performed by the cloud computing task scheduling platform that connects to the master node through the network.
  • FIG. 14 is a side view of a rack in a customized rack structure according to an embodiment of the present invention.
  • the ring slot layer of the rack may be a multi-layer design, and the arrowed line indicates the running path of the backup hardware.
  • the hardware slot is unlocked, and the damaged hardware is disconnected from the rack due to centrifugal force; the backup hardware runs through the conveyor to the backup guide; the ring gradually reduces the speed until it stops rotating, so that the backup guide The slot and the hardware need to be replaced by the slot, and then the backup device is loaded into the hardware slot through the operation of the transmitting device; after the replacement hardware is replaced, the ring gradually returns to the original running rotation speed.
  • FIG. 16 is a schematic flowchart of an embodiment of a mobile cloud computing intermediate platform application according to an embodiment of the present invention.
  • the service flow between two master devices is for example, the business process between the Master and other Slave nodes and the industry shown in Figure 16. The same as the business process, it will not be repeated here. As shown in Figure 16, it includes:
  • Step 1600 Install a mobile cloud computing intermediate platform, Hadoop, and a game application on the Master node and the Slave node, respectively.
  • Step 1601 The related configuration of the Android system and the Hadoop platform is implemented and implemented by using the mobile cloud computing intermediate platform, and the method generally includes:
  • the mobile cloud computing intermediate platform of the master node provides the Java library missing from the Android system for the Hadoop platform; the mobile cloud computing intermediate platform of the master node performs hardware selection and loading on the Android system; the mobile cloud computing intermediate platform of the master node allocates an IP address, according to The different network connection methods are used to configure the corresponding IP address; the Android system of the intelligent terminal reads the assigned IP address and writes it to the Hadoop configuration file and takes effect; the mobile cloud computing intermediate platform of the master node performs the data resource migration in the cluster in the Hadoop platform. And computing resource integration, and notify other Slave nodes in the cluster to make resource adjustments.
  • the implementation is shown in Figure 2 and will not be described here.
  • Step 1602 One or more nodes, as shown in Figure 16, the Master node and the Slave node detect the resource usage of the game application, and feed back to the respective Hadoop platform; the Hadoop platform of one or more Slave nodes receives the resource occupied by the game application. , feedback to the Master node.
  • Step 1603 The master node automatically performs data resource migration and computing resource integration according to one or more node storage and computing resource conditions.
  • Step 1604 The game application starts to be used. After receiving the usage behavior and the task, the master node allocates calculation and storage tasks to one or more slave nodes; one or more slave nodes feed back the calculation result and the storage result to the master node, and the master node performs The result of the calculation and the stored result are summarized, and the game application operation result and the storage result are fed back to the game application.
  • Step 1605 During the calculation task and the storage task of the game application, the mobile cloud computing intermediate platform of the master node performs at least one of the following:
  • Step 1606 The game application ends, and the Master node notifies one or more Slave nodes to release computing resources and storage resources; one or more Slave nodes feed back the computing resources and storage resource release results to the Master node.
  • the embodiment of the invention further provides a computer readable storage medium storing computer executable instructions, which are implemented by the processor to implement the method described in the foregoing embodiments.
  • computer storage medium includes volatile and nonvolatile, implemented in any method or technology for storing information, such as computer readable instructions, data structures, program modules, or other data. , removable and non-removable media.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridge, magnetic tape, magnetic disk storage or other magnetic storage device, or may Any other medium used to store the desired information and that can be accessed by the computer.
  • communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and can include any information delivery media.
  • the mobile cloud computing intermediate platform is established between the Hadoop and the Android system, and is used for ensuring the adaptation and operation between the Hadoop and the Android system, so that the Hadoop implements the normal operation on the Android system, that is, the Android system supports. Hadoop runs.
  • the intelligent terminal realizes a very convenient distributed system for building, joining, and removing a wireless network or a wired network, and particularly realizes that the use is cheaper and more convenient to replace. Intelligent terminals form a distributed cluster.

Abstract

一种实现移动云计算中间平台的方法及实现分布式的方法，其中，实现移动云计算中间平台的方法包括：进行网络配置；根据加入分布式系统的智能终端设置Hadoop平台的运行配置（S1701）；对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理（S1702）；根据网络配置的信息及整合后的计算和存储资源进行云计算管理（S1703）；所述移动云计算中间平台建立在Hadoop平台和安卓Android系统之间。

Description

实现移动云计算中间平台的方法及实现分布式的方法 技术领域
本文涉及但不限于大数据技术，尤指一种实现移动云计算中间平台的方法及实现分布式的方法。
背景技术
Hadoop是一个由Apache基金会所开发的分布式系统基础架构。用户可以在不了解分布式底层细节的情况下，开发分布式程序。充分利用集群的威力进行高速运算和存储。Hadoop分布式文件系统(HDFS，Hadoop Distributed File System)具有高容错性的特点，并设计用于部署在相比小型机更为低廉的(low-cost)PC服务器上；而且HDFS能提供高吞吐量(high throughput)来访问应用程序的数据，适合那些有着超大数据集(large data set)的应用程序。HDFS放宽(relax)了POSIX的要求，可以以流的形式访问(streaming access)文件系统中的数据。
Hadoop框架最核心的设计包括：HDFS和MapReduce。HDFS为海量的数据提供了存储，而MapReduce为海量的数据提供了计算。目前，Hadoop开源免费、硬件廉价、开发便利，因此，在大数据的通行市场上占据主导地位，使得Hadoop市场份额逐年高速递增。
实际上，在软件的成本之外，硬件产生的经济成本更为可观。如何降低硬件成本是每一个服务器制造商都在考虑并寻求解决方法的课题。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提供一种实现移动云计算中间平台的方法及实现分布式的方法，能够使得Android系统支持Hadoop的运行。
本发明实施例提供了一种实现移动云计算中间平台的方法，包括：
进行网络配置；根据加入分布式系统的智能终端设置Hadoop平台的运行配置；
对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理；
根据网络配置的信息及整合后的计算和存储资源进行云计算管理；
所述移动云计算中间平台建立在Hadoop平台和安卓Android系统之间。
可选地，还包括：为所述Hadoop平台提供所述Android系统缺少的Java库。
可选地，所述网络配置包括：
所述移动云计算中间平台第一次启动时，为安装所述移动云计算中间平台的智能终端分配一个固定地址；
根据该智能终端所属分布式系统的连接方式获取IP地址。
可选地，所述根据智能终端所属分布式系统的连接方式获取IP地址包括：
所述连接方式为有线连接时，搜寻所述分布式系统的所有智能终端的IP地址，对最大的IP地址进行处理后作为所述IP地址；
所述连接方式为无线连接时，所述IP地址由无线网动态主机配置协议DHCP自动分配；
所述连接方式为定制机架方式时，根据槽位的不同来获取所述IP地址。
可选地，所述对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理包括：
当安装所述移动云计算中间平台的智能终端所属分布式系统接收到写入和计算任务时，确认当前分布式系统中的资源使用状况；
根据确认的资源使用状况调取所述Hadoop平台的任务分配方法：当存储大小超出设定限额上限的不接受写入操作，当计算能力超出设定限额上限的不接受计算任务；当存储大小低于设定限额下限时，优先写入，直到达到 设定限额的上限；当计算能力低于设定限额下限时，优先接受计算任务，直到达到设定限额的上限；所述分布式系统中存储大小和计算能力处于上限和下限之间的智能终端获取任务时进行随机分配。
可选地，还包括：为安装所述移动云计算中间平台的智能终端所述分布式系统中的智能终端提供硬件健康监控，并提供修复操作。
可选地，当所述硬件出现问题后，还包括：进行重连或重启操作，预设次数重连或重启仍失败后，提示替换所述硬件。
可选地，还包括：当系统日志大于设定的范围值时，进行先入先出FIFO存储。
本发明实施例还提供了一种用于实现移动云计算中间平台的装置，至少包括存储器和处理器，其中，存储器中存储有以下可执行指令：
进行网络配置；根据加入分布式系统的智能终端设置Hadoop平台的运行配置；对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理；根据网络配置的信息及整合后的计算和存储资源进行云计算管理；
所述移动云计算中间平台建立在Hadoop平台和安卓Android系统之间。
本发明实施例还提供了一种实现分布式的方法，包括：
获取并安装移动云计算中间平台和Hadoop平台到智能终端；
移动云计算中间平台在Hadoop平台中进行所述智能终端所属分布式系统内数据资源迁移以及计算资源整合；
移动云计算中间平台接收来自Hadoop平台反馈的所述分布式系统中的智能终端的存储和计算资源情况，并进行云计算管理。
可选地，当有新的智能终端接入所述分布式系统时，还包括：
所述移动云计算中间平台根据新接入的智能终端的存储和计算资源情况，重新进行数据资源迁移以及计算资源整合。
可选地，当有新的智能终端接入所述分布式系统时，还包括：
新的智能终端连接至所述分布式系统所在网络，获取并安装所述移动云计算中间平台及Hadoop平台；
新的智能终端使用安装的移动云计算中间平台，按照所述分布式系统所在网络的连接方式配置Hadoop系统网络。
可选地，还包括根据智能终端所属分布式系统的连接方式获取IP地址，包括：
所述连接方式为有线连接时，搜寻所述分布式系统的所有智能终端的IP地址，对最大的IP地址进行处理后作为所述IP地址；
所述连接方式为无线连接时，所述IP地址由无线网动态主机配置协议DHCP自动分配；
所述连接方式为定制机架方式时，根据槽位的不同来获取所述IP地址。
可选地，所述获取并安装移动云计算中间平台和Hadoop平台之后，所述进行分布式系统内数据资源迁移以及计算资源整合之前，还包括：
所述移动云计算中间平台提供所述智能终端的Android系统缺少的Java库，对所述智能终端的Android系统进行硬件选取、加载；
所述移动云计算中间平台根据智能终端返回的分布式系统连接方式为所述智能终端分配IP地址；并写入Hadoop平台的配置文件。
可选地，所述进行数据资源迁移以及计算资源整合包括：
存储大小超出设定限额上限的智能终端不接受写入操作，计算能力超出设定限额上限的职能终端不接受计算任务；
当智能终端的存储大小低于设定限额下限时，优先写入，直到达到设定限额的上限；当智能终端的计算能力低于设定限额下限时，优先接受计算任务，直到达到设定限额的上限；
存储大小和计算能力在上限和下限之间的智能终端获取任务时，进行随机分配。
可选地，还包括：所述移动云计算中间平台采用FIFO存储方式存储日志。
可选地，还包括：所述移动云计算中间平台根据所述Hadoop平台反馈的所述分布式系统接入的智能终端的心跳情况进行硬件健康状况的监控。
可选地，所述移动云计算中间平台提供给外部接口进行云计算管理。
可选地，当所述分布式系统中有智能终端退出所述分布式系统时，还包括：
所述移动云计算中间平台接收到来自所述Hadoop平台的退出分布式系统申请，在所述Hadoop平台中再次进行所述分布式系统内数据资源迁移以及计算资源整合；
在所述智能终端退出成功后，所述移动云计算中间平台记录日志。
本发明实施例移动云计算中间平台建立在Hadoop和Android系统之间，用于保障Hadoop和Android系统之间的适配以及运行，使得Hadoop实现了在Android系统上的正常运行即使得Android系统支持了Hadoop的运行。另一方面，通过本发明实施例移动云计算中间平台，智能终端实现了非常方便的组建、加入、移出无线网络或有线网络组建的分布式系统，特别地实现了使用更为廉价且方便更换的智能终端组建分布式集群。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
可选图1为本发明实施例移动云计算中间平台的组成结构示意图；
图2为本发明实施例移动云计算中间平台的工作原理流程示意图；
图3为本发明实施例基于Android系统的HDFS集群实现数据存储的流程示意图；
图4为本发明实施例基于Android系统的HDFS集群实现计算的流程示意图；
图5为本发明实施例改造后的智能终端中计算模块的组成结构示意图；
图6为本发明实施例改造后的智能终端中存储模块的组成结构示意图；
图7为本发明实施例智能终端采用WIFI连接方式的组网示意图；
图8为本发明实施例基于图7所示的连接方式的对智能终端在集群中的管理的实施例的流程示意图；
图9为本发明实施例智能终端采用移动网络连接方式的组网示意图；
图10为本发明实施例基于图9所示的连接方式的对智能终端在集群中的管理的实施例的流程示意图；
图11为本发明实施例智能终端采用有线连接方式的组网示意图；
图12为本发明实施例基于图11所示的连接方式的对智能终端在集群中的管理的实施例的流程示意图；
图13为本发明实施例定制机架结构中机架的俯视图；
图14为本发明实施例定制机架结构中机架侧视图；
图15为本发明实施例定制机架结构中实现硬件替换的示意图；
图16为本发明实施例移动云计算中间平台应用的实施例的流程示意图；
图17为本发明实施例实现移动云计算中间平台的方法流程图；
图18为本发明实施例建立的移动云计算中间平台实现分布式的方法流程图。
本发明的实施方式
下文中将结合附图对本发明的实施例进行详细说明。
目前，英特尔(Intel)公司的中央处理单元(CPU，Central Processing Unit)占据着市场的垄断地位，导致PC服务器的CPU价格居高不下。而智能终端所用的CPU处理器价格低廉。这两者价格相差30倍左右，功耗相差1000倍左右。而两者之间的性能差距，平均到单核上，中央处理器只有移动终端(如手机)处理器的5倍左右。一台100个智能终端组成的集群(如手机集群)，成本可以控制在1到2万元，物理体积和一台普通PC服务器相当，计算性能是一台PC服务器的5-10倍，功耗可以控制在服务器的1/10到1/20。
目前，基于智能终端的Android系统不能支持Hadoop的运行。另外，现有智能终端要组建集群，除了通过无线信号(移动网络、WIFI网络)建立之外，也可以对智能终端的硬件进行改造，如增设有线通信模块。这种组 件集群的方式，主要依赖于硬件的实现，实现繁琐复杂，而且很难对成本进行控制。
图1为本发明实施例移动云计算中间平台的组成结构示意图，如图1所示，本发明实施例移动云计算中间平台是建立在Hadoop平台和Android系统之间的一个中间层，用于保障Hadoop平台和Android系统之间的适配以及运行。至少包括：硬件驱动模块、Java库支撑模块、配置模块、计算存储资源管理模块、云计算任务调度模块；其中，
硬件驱动模块，设置为为操作系统加载硬件驱动。该驱动主要为网络控制器服务，比如：当通信为RJ45接口时，网络控制器为USB转RJ45信号芯片；当通信为光纤通信时，网络控制器为光电转换芯片。
Java库支撑模块，设置为为Hadoop平台提供Android系统缺少的Java库。Android系统本身支持一部分Java功能，这里的Java库支撑模块用于补足Android系统中缺少的Java库。
配置模块，设置为进行网络配置；根据加入分布式系统的智能终端如该智能终端的计算能力、存储大小自动设置Hadoop平台的运行配置。
其中，配置模块是设置为通过如下方式实现网络配置：
配置模块自身所在移动云计算中间平台在第一次启动时，为安装所述移动云计算中间平台的智能终端分配一个固定地址(所有机器初次启动时都为该IP地址)；
网络接通后确定该智能终端所属分布式系统的连接方式如有线、无线或者定制机架类型等，根据连接方式获取IP地址：当为有线连接时，搜寻当前网络的所有机器的IP地址，然后对最大的IP地址进行处理：在最大的IP地址第四段+1，当加到253时，则在IP地址第三段加1，第四段从1开始计数；当为无线连接时，IP地址由无线网动态主机配置协议(DHCP，Dynamic Host Configuration Protocol)自动分配，不需要考虑IP地址分配，只需要读取分配的IP地址并写入Hadoop配置文件中即可；当为定制机架方式时，智能终端根据槽位的不同来获取IP地址配置，这样，使得新替换上电的设备配置和出现问题的设备配置相同，即同样的槽位的IP地址是不可变更的。
计算存储资源管理模块，设置为在现有Hadoop计算和存储控制的基础上，整合到计算存储资源管理模块自身所在移动云计算中间平台中，对计算和存储的资源进行管理，基于计算和存储资源来进行整合处理。实现包括：当分布式系统接收外来写入和计算任务时，先确认当前分布式系统中机器的资源使用状况，然后调取Hadoop平台的任务分配方法：存储已经超出设定限额上限的将不接受写入操作，计算超出设定限额上限的将不接受计算任务；当存储低于设定限额下限时将优先写入，直到达到设定限额的上限；当计算低于设定限额下限时将优先接受计算任务，直到达到设定限额的上限；其他处于上限和下限之间的机器获取任务时进行随机分配。
云计算任务调度模块，设置为提供给外部接口，根据网络配置的信息及整合后的计算和存储资源进行云计算管理，管理包括启动、停止、暂停、定时等。
可选地，本发明实施例移动云计算中间平台还包括：
健康监控模块，设置为为智能终端提供硬件健康监控，并提供修复操作。需要说明的是，健康监控模块只监控硬件死活，当硬件出现问题后会进行重连或重启操作，多次重连或重启失败后，提示替换该硬件。监控功能可以使用心跳检测机制来监控硬件死活，使用温度监控来对硬件进行运行健康告警等。
可选地，本发明实施例移动云计算中间平台还包括：
日志处理模块，设置为当系统日志大于设定的范围值时，日志的存储进行先入先出(FIFO，First in first out)存储，即先删除远期日志，然后存储最新的日志。
图1只列出了本发明实施例移动云计算中间平台的基本的Hadoop组件，其他组件可以自行选择安装，是否包括其他组件、包括多少，并不用于限定本发明的保护范围，这里不再赘述。
相应地，本发明实施例还提供一种实现移动云计算中间平台的方法，如图17所示，包括：将移动云计算中间平台建立在Hadoop平台和安卓Android系统之间，还包括：
S1701进行网络配置；根据加入分布式系统的智能终端设置Hadoop平 台的运行配置；
S1702对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理；
S1703根据网络配置的信息及整合后的计算和存储资源进行云计算管理。
本发明实施例方法还包括：为Hadoop平台提供Android系统缺少的Java库。
可选地，进行的网络配置包括：
在建立的移动云计算中间平台第一次启动时，为安装所述移动云计算中间平台的智能终端分配一个固定地址；根据该智能终端所属分布式系统的连接方式获取IP地址。其中，
根据智能终端所属分布式系统的连接方式获取IP地址包括：
当连接方式为有线连接时，搜寻分布式系统的所有智能终端的IP地址，对最大的IP地址进行处理后作为所述IP地址；
当连接方式为无线连接时，IP地址由无线网DHCP自动分配；
当连接方式为定制机架方式时，根据槽位的不同来获取所述IP地址。
可选地，对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理包括：
当安装移动云计算中间平台的智能终端所属分布式系统接收到写入和计算任务时，确认当前分布式系统中的资源使用状况；
根据确认的资源使用状况调取Hadoop平台的任务分配方法：当存储大小超出设定限额上限的不接受写入操作，当计算能力超出设定限额上限的不接受计算任务；当存储大小低于设定限额下限时，优先写入，直到达到设定限额的上限；当计算能力低于设定限额下限时，优先接受计算任务，直到达到设定限额的上限；所述分布式系统中存储大小和计算能力处于上限和下限之间的智能终端获取任务时进行随机分配。
可选地，本发明实施例实现移动云计算中间平台的方法还包括：
为安装移动云计算中间平台的智能终端所述分布式系统中的智能终端提供硬件健康监控，并提供修复操作。
当硬件出现问题后，还包括：进行重连或重启操作，预设次数重连或重启仍失败后，提示替换该硬件。
可选地，本发明实施例实现移动云计算中间平台的方法还包括：当系统日志大于设定的范围值时，进行FIFO存储。
本发明实施例还提供一种用于实现移动云计算中间平台的装置，至少包括存储器和处理器，其中，
存储器中存储有以下可执行指令：将移动云计算中间平台建立在Hadoop平台和安卓Android系统之间；进行网络配置；根据加入分布式系统的智能终端设置Hadoop平台的运行配置；对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理；根据网络配置的信息及整合后的计算和存储资源进行云计算管理。
如图18所示，利用本发明实施例建立的移动云计算中间平台，实现分布式的方法包括：
S1801获取并安装移动云计算中间平台和Hadoop平台到智能终端；
S1802移动云计算中间平台在Hadoop平台中进行所述智能终端所属分布式系统内数据资源迁移以及计算资源整合；
S1803移动云计算中间平台接收来自Hadoop平台反馈的所述分布式系统中的智能终端的存储和计算资源情况，并进行云计算管理。
下面结合一台智能终端在安装本发明实施例图1所示的移动云计算中间平台后各模块的工作流程情况为例，对本发明实施例移动云计算中间平台的工作原理进行详细描述。图2为本发明实施例移动云计算中间平台的工作原理流程示意图，如图2所示，包括：
首先，智能终端获取并安装本发明实施例移动云计算中间平台和Hadoop平台；本发明实施例移动云计算中间平台的Java库支撑模块提供智能终端的Android系统缺少的Java库；本发明实施例移动云计算中间平台的硬件驱动模块对智能终端的Android系统进行硬件选取、加载；本发明实施 例移动云计算中间平台的配置模块为智能终端分配一个IP地址：智能终端会向本发明实施例移动云计算中间平台的配置模块返回分布式系统连接类型；本发明实施例移动云计算中间平台的配置模块根据智能终端的不同网络连接方式进行相应IP地址的配置：当为有线连接时，搜寻当前网络的所有机器的IP地址，然后对最大的IP地址进行处理：在最大的IP地址第四段+1，当加到253时，则在IP地址第三段加1，第四段从1开始计数；当为无线连接时，IP地址由无线网DHCP自动分配；当为定制机架方式时，根据槽位的不同来获取IP地址配置。智能终端读取分配的IP地址写入Hadoop平台的配置文件并生效；
然后，本发明实施例移动云计算中间平台的计算存储资源管理模块在Hadoop平台中进行分布式系统内数据资源迁移以及计算资源整合，Hadoop平台将智能终端的存储和计算资源情况反馈给本发明实施例移动云计算中间平台的计算存储资源管理模块；
接着，本发明实施例移动云计算中间平台的计算存储资源管理模块根据新接入的智能终端的存储和计算资源情况，进行数据资源迁移以及计算资源整合：存储已经超出设定限额上限的将不接受写入操作，计算超出设定限额上限的将不接受计算任务；当存储低于设定限额下限时将优先写入，直到达到设定限额的上限；当计算低于设定限额下限时将优先接受计算任务，直到达到设定限额的上限；其他处于上限和下限之间的机器获取任务时进行随机分配；
最后，对于日志的存储，本发明实施例移动云计算中间平台的日志处理模块进行FIFO存储；本发明实施例移动云计算中间平台的健康监控模块进行硬件健康的监控，Hadoop平台反馈接入的智能终端的心跳情况以检测硬件死活，反馈接入的智能终端的节点温度以监控运行健康状况；本发明实施例移动云计算中间平台的云计算任务调度模块提供给外部接口进行云计算管理。
后续，如果智能终端申请退出分布式系统，Hadoop平台会向本发明实施例移动云计算中间平台的计算存储资源管理模块申请退出分布式系统，本发明实施例移动云计算中间平台的计算存储资源管理模块会在Hadoop平台 中再次进行分布式系统内数据资源迁移以及计算资源整合，并在退出成功后，在本发明实施例移动云计算中间平台的日志处理模块中记录日志。
综上所述，通过本发明实施例基于Android系统的Hadoop平台即图1所示的移动云计算中间平台，一方面，使得Hadoop平台实现了在Android系统上的正常运即使得Android系统支持了Hadoop平台的运行。可选地，提供了Hadoop平台运行时软件和硬件的健康监控，从而保证了在损坏时触发对软件的修复或对硬件的替换。另一方面，通过本发明实施例基于Android系统的Hadoop平台即图1所示的移动云计算中间平台，智能终端实现了非常方便的组建、加入、移出无线网络或有线网络组建的分布式系统。从而实现了使用更为廉价且方便更换的智能终端组建分布式系统。
本发明实施例以智能终端为手机终端进行举例描述，但是，本发明实施例智能终端包括：运行Android系统的手机终端、PAD、机顶盒，以及运行Android系统的每类有线、无线手持终端等。
为方便描述，本发明实施例中以分布式系统为集群为例进行描述，但并不用于限定本发明的保护范围。
需要说明一下，Hadoop平台通过本发明实施例移动云计算中间平台对底层Android系统调用一个或多个应用接口(API)功能进行数据读写、计算等任务。Android API接口的封装包括：
对于智能终端内存数据读写，可以包括：
getFileDir()：用于获取Hadoop平台在智能终端内存存储数据的位置信息，如：/data/data/Hadoop/files；
getCacheDir()：用于获取Hadoop平台在智能终端内存缓存数据的位置信息，如：/data/data/Hadoop/cache；
openFileInput(String name)：用于直接获取/data/data/Hadoop/files/name文件的输入流信息；
openFileOutput(String name,int mode)：用于直接获取/data/data/Hadoop/files/name文件的输出流，其中，mode为写入文件时的权限；
Context.MODE_PRIVATE：为私有模式(或称为默认模式)，只能被应用本身和同一群组的智能终端访问；写入的内容覆盖原文件内容；
Context.MODE_APPEND：为追加模式(或称为私有模式)，只能被应用本身和同一群组的智能终端访问；如果文件存在就追加内容，如果文件不存在就新建文件并写入内容；
Context.MODE_WORLD_READABLE：为群组中所有智能终端可读权限；
Context.MODE_WORLD_WRITEABLE：为群组中所有智能终端可写权限；
对于SDcard数据读写，可以包括：
getExternalStorageDirectory()：用于获取Hadoop平台所在智能终端的SDcard位置信息，如/storage/sdcard；
getExternalStorageState()：用于获取Hadoop平台所在智能终端的SDcard的当前状态，比较常用的应该是MEDIA_MOUNTED；
FileInputStream()，用于读取文件；
BufferReader()，用于读取文件；
httpConnection()，用于将读取的流保存为String数据；
FileOutputStream()，用于写入文件；
BufferedWriter()，用于写入文件；
另外，Hadoop计算使用Java库的加、减、乘、除四则运算。
本发明实施例中，基于Android系统的HDFS集群有两类节点并以“管理者—工作者”模式运行：一个作为管理者的主(Master)节点(NameNode)和多个作为工作者的从(Slave)节点(DataNode)。每一个NameNode和DataNode都对应一台智能终端。
其中，NameNode，主要负责管理HDFS文件系统，包括namespace管理、块(block)管理；DataNode，主要设置为存储数据文件。这里，HDFS会将一个文件分割成一个个的block，这些block可能存储在一个DataNode 上或者是多个DataNode上。DataNode负责实际的底层文件的读写，如果客户端(Client)程序发起了读HDFS上的文件的命令，那么，首先会将这些文件分成若干个block，然后NameNode将告知Client这些block数据是存储在哪些DataNode上的，这样，Client直接和DataNode交互即可。
下面详细介绍本发明实施例基于Android系统的HDFS集群的存储和计算过程。
图3为本发明实施例基于Android系统的HDFS集群实现数据存储的流程示意图，以文件上传到分布式集群进行存储为例，这里，NameNode负责管理存储在HDFS上所有文件的元数据，确认客户端的请求，并记录文件的名字和存储该文件的数据节点即DataNode的集合。并将这些记录的信息存储在内存中的文件分配表中。本实施例中，假设客户端发送一个请求给NameNode，表明要将“swim.txt”文件写入到HDFS，如图3所示，包括：
步骤1：客户端发消息给NameNode，表明要写入“swim.txt”文件；
步骤2：NameNode向客户端返回消息，通知客户端将“swim.txt”文件写到DataNode A、DataNode B和DataNode D，并直接与DataNode B联系；
步骤3：客户端发消息给DataNode B，指示DataNode B保存一份“swim.txt”文件，并分别发送一份副本给DataNode A和DataNode D；
步骤4：DataNode B发消息给DataNode A，指示DataNode A保存一份“swim.txt”文件，并发送一份副本给DataNode D；
步骤5：DataNode A发消息给DataNode D，指示DataNode D保存一份“swim.txt”文件；
步骤6：DataNode D发确认消息给DataNode A；
步骤7：DataNode A发确认消息给DataNode B；
步骤8：DataNode B发确认消息给客户端，表示完成“swim.txt”文件的写入。
这样，一份“swim.txt”文件就保存在了分布式集群的DataNode A、DataNode B和DataNode D三个数据节点(即智能终端)上。
图4为本发明实施例基于Android系统的HDFS集群实现计算的流程示意图，如图4所示，在基于Android的HDFS中，每个MapReduce任务都被初始化为一个Job，每个Job又可以分为两种阶段：映射(Map)阶段和归约(Reduce)阶段。这两个阶段分别用两个函数表示，即Map函数和Reduce函数。其中，Map函数接收一个<key,value>形式的输入，然后同样产生一个<key,value>形式的中间输出；Reduce函数接收一个如<key,(list of values)>形式的输入，然后对这个value集合进行处理，每个Reduce产生0或1个输出，Reduce的输出也是<key,value>形式的。
一般，常见的智能终端在安装本发明实施例的移动云计算中间平台和Hadoop平台后，使用稳定的无线连接，经过本发明实施例的移动云计算中间平台对Hadoop平台进行配置后，即可进行Hadoop之上的一个或多个组件的使用。比如在办公室中，可以将大家的智能终端集中到一个集群内，实现计算资源和存储资源共享。这样，一个人的智能终端无法做到的计算量，通过十几个人的智能终端就可以组建起集群，并在本发明实施例的移动云计算中间平台纸上完成本来不能完成的计算量。无线连接又包括：WIFI网络、移动网络(如2G、3G、4G等)两种集群组建方式。
为了保障集群的可靠性、增强通信速度，可以使用有线连接方式。使用有线连接方式的集群硬件来源可以为定制的智能终端，也可以使用现有智能终端进行有线通信的改造。对现有智能终端的改造主要包括两部分：一方面是，对USB接口进行改造，将USB类型的通信转换为双绞线或者光纤线通信；另一方面，存储模块至少包括ROM、SD卡以及接口。存储模块损坏是可以替换的。
图5为本发明实施例改造后的智能终端中计算模块的组成结构示意图，如图5所示，至少包括：中央处理器单元(CPU，Central Processing Unit)芯片组、随机存储器(RAM，Random-Access Memory)、网络控制器。对外包括电源、网口、存储接口3个接口。
其中，USB接口可以包括：USB2.0或USB3.0，分别对应双绞线和光纤线使用。因此，针对USB的两种不同类型的通信方案。网络控制器采用常见的USB转网卡的方案，如果是光纤再添加光电转换芯片即可。图5中的 网络控制器是一个信号转换芯片，比如：USB2.0转RJ45信号芯片如AX88772B；USB 3.0-to-Gigabit Ethernet控制芯片如RTL8153等。
图5中，安全数据(SD，Secure Digital)卡接口和Android系统定制系统镜像只读存储器(ROM，Read Only Memory Image)接口合并为存储接口，只是硬件外观的变化，实际上两个接口还是独立存在的。
图6为本发明实施例改造后的智能终端中存储模块的组成结构示意图，如图6所示，至少包括内置两个组件：ROM和SD卡，对外显示为一个存储接口。该模块损坏可以进行替换。
Hadoop是用途广泛的大数据平台，目前基于Hadoop的在用实施例都可以使用本发明实施例提供的技术方案，由PC服务器转移到安装有Android系统的智能终端中。下面根据智能终端的不同网络连接方式，结合本发明实施例移动云计算中间平台对智能终端在集群中的管理的实施例进行详细描述。
图7为本发明实施例智能终端采用WIFI连接方式的组网示意图，图8为本发明实施例基于图7所示的连接方式的对智能终端在集群中的管理的实施例的流程示意图，结合图7和图8，本实施例以新的节点为智能终端为例，将智能终端加入集群的步骤包括：
步骤800：将欲加入现有集群的WIFI网络的智能终端成功连接WIFI网络。
步骤801：欲加入现有集群的WIFI网络的智能终端通过自身Android系统下载获取并安装移动云计算中间平台及Hadoop平台。
步骤802：使用移动云计算中间平台配置Hadoop平台系统网络。实现过程如图2所示，这里不再赘述。
假设连接方式的连接(Connection)参数为1，表示为WIFI连接方式，Connection参数为2，表示为移动网络如2G、3G、4G连接方式，Connection参数为3，表示为有线连接方式。
本实施例中，假设Connection参数为1，则移动云计算中间平台根据该WIFI网络中的DHCP机制给本实施例中新加入的节点(如步骤800中的智能终端)分配IP地址；并将分配的IP地址写入集群配置文件中。
步骤803：现有集群中的主节点启动Hadoop集群组件并成功启动。
步骤804：现有集群中的主节点进行集群内数据资源迁移以及计算资源整合(balance)。
本实施例中，主节点中的移动云计算中间平台根据新节点(如智能终端)的存储和计算资源情况，自行进行数据资源迁移以及计算资源整合。举例来看，假设每个节点的存储资源上限为总存储资源的80％以及存储资源下限为总存储资源的5％；假设每个节点的CPU计算资源上限为CPU利用率达到80％以及CPU计算资源下限为CPU利用率达到5％；其中，上限、下限的设置可以按照实际需求进行调整。
移动云计算中间平台根据设置的上限、下限值进行如下操作：
当该新加入的节点的存储已经超出存储资源上限时，将不接受写入操作；
当该新加入的节点的计算超出CPU计算资源上限时，将不接受计算任务；
当该新加入的节点的存储低于存储资源下限时，将优先写入，直到达到存储资源上限；
当该新加入的节点的计算低于CPU计算资源下限时，将优先接受计算任务，直到达到CPU计算资源上限；
其他处于上限和下限之间的情况，获取任务时可以进行随机分配。
步骤805：现有集群中的主节点调用移动云计算中间平台接口，提交Hadoop任务进行运行。
步骤806：集群中的数据节点完成任务计算，并将任务结果反馈给主节点。
本实施例中，假设某个数据节点如步骤800中加入现有集群的智能终端申请退出集群，包括以下步骤：
步骤807：该智能终端向集群中的主节点申请退出集群。
步骤808：主节点进行数据资源迁移以及计算资源整合。实现如步骤804所述，这里不再赘述。
步骤809：资源整合完成后，同意该智能终端退出集群。
图9为本发明实施例智能终端采用移动网络连接方式的组网示意图，图10为本发明实施例基于图9所示的连接方式的对智能终端在集群中的管理的实施例的流程示意图，结合图9和图10，本实施例以新的节点为智能终端为例，将智能终端加入集群的步骤包括：
步骤1000：将欲加入现有集群的移动网络如2G、3G、4G等的智能终端成功连接移动网络。
步骤1001：欲加入现有集群的移动网络的智能终端通过自身Android系统下载获取并安装移动云计算中间平台及Hadoop平台。
步骤1002：使用移动云计算中间平台配置Hadoop系统网络。实现过程如图2所示，这里不再赘述。
假设连接方式的连接(Connection)参数为1，表示为WIFI连接方式，Connection参数为2，表示为移动网络如2G、3G、4G连接方式，Connection参数为3，表示为有线连接方式。
本实施例中，假设Connection参数为2，则移动云计算中间平台根据该移动网络中的DHCP机制给本实施例中新加入的节点(如步骤1000中的智能终端)分配IP地址；并将分配的IP地址写入集群配置文件中。
步骤1003：现有集群中的主节点启动Hadoop集群组件并成功启动。
步骤1004：现有集群中的主节点进行集群内数据资源迁移以及计算资源整合(balance)。
本实施例中，主节点中的移动云计算中间平台根据新节点(如智能终端)的存储和计算资源情况，自行进行数据资源迁移以及计算资源整合。举例来看，假设每个节点的存储资源上限为总存储资源的80％以及存储资源下限为总存储资源的5％；假设每个节点的CPU计算资源上限为CPU利用率达到 80％以及CPU计算资源下限为CPU利用率达到5％；其中，上限、下限的设置可以按照实际需求进行调整。
移动云计算中间平台根据设置的上限、下限值进行如下操作：
当该新加入的节点的存储已经超出存储资源上限时，将不接受写入操作；
当该新加入的节点的计算超出CPU计算资源上限时，将不接受计算任务；
当该新加入的节点的存储低于存储资源下限时，将优先写入，直到达到存储资源上限；
当该新加入的节点的计算低于CPU计算资源下限时，将优先接受计算任务，直到达到CPU计算资源上限；
其他处于上限和下限之间的情况，获取任务时可以进行随机分配。
可选地，考虑到传输速率以及流量资费的因素，本实施例中还可设置数据压缩(Compress)参数，当Compress参数为1时，表示需要进行压缩，当Compress参数为0时，表示不需要进行压缩。
当Connection参数为2时，即本实施例中采用移动网络的方式，Compress参数自动设置为1，表示数据传输时需要进行压缩，以获得高传输速率，从而降低流量资费；
当Connection参数为1或3时，Compress参数自动设置为0，数据传输不需要进行压缩，以减轻CPU运算压力。
步骤1005：现有集群中的主节点调用移动云计算中间平台接口，提交Hadoop任务进行运行。
步骤1006：集群中的数据节点完成任务计算，并将任务结果反馈给主节点。
本实施例中，假设某个数据节点如步骤1000中加入现有集群的智能终端申请退出集群，包括以下步骤：
步骤1007：该智能终端向集群中的主节点申请退出集群。
步骤1008：主节点进行数据资源迁移以及计算资源整合。实现如步骤1004所述，这里不再赘述。
步骤1009：资源整合完成后，同意该智能终端退出集群。
图11为本发明实施例智能终端采用有线连接方式的组网示意图，图12为本发明实施例基于图11所示的连接方式的对智能终端在集群中的管理的实施例的流程示意图，结合图11和图12，本实施例以新的节点为智能终端为例，将智能终端加入集群的步骤包括：
步骤1200：将欲加入现有集群的的智能终端使用双绞线或光纤成功连接集群中的交换机。本实施例中，欲加入现有集群的智能终端使用USB2.0接口，因此，采用双绞线连接集群的交换机。
步骤1201：欲加入现有集群的移动网络的智能终端通过自身Android系统下载获取并安装移动云计算中间平台及Hadoop平台。
步骤1202：使用移动云计算中间平台配置Hadoop系统网络。实现过程如图2所示，这里不再赘述。
假设连接方式的连接(Connection)参数为1，表示为WIFI连接方式，Connection参数为2，表示为移动网络如2G、3G、4G连接方式，Connection参数为3，表示为有线连接方式。
本实施例中，假设Connection参数为3，则移动云计算中间平台搜寻当前集群网络所有节点的IP地址，然后对最大的IP地址进行处理：在最大的IP地址第四段+1，当加到253时，则在IP地址第三段加1，第四段从1开始计数；并将分配的IP地址写入集群配置文件中。
步骤1203：现有集群中的主节点启动Hadoop集群组件并成功启动。
步骤1204：现有集群中的主节点进行集群内数据资源迁移以及计算资源整合(balance)。
本实施例中，主节点中的移动云计算中间平台根据新节点(如智能终端)的存储和计算资源情况，自行进行数据资源迁移以及计算资源整合。举例来看，假设每个节点的存储资源上限为总存储资源的80％以及存储资源下限为总存储资源的5％；假设每个节点的CPU计算资源上限为CPU利用率达到 80％以及CPU计算资源下限为CPU利用率达到5％；其中，上限、下限的设置可以按照实际需求进行调整。
移动云计算中间平台根据设置的上限、下限值进行如下操作：
当该新加入的节点的存储已经超出存储资源上限时，将不接受写入操作；
当该新加入的节点的计算超出CPU计算资源上限时，将不接受计算任务；
当该新加入的节点的存储低于存储资源下限时，将优先写入，直到达到存储资源上限；
当该新加入的节点的计算低于CPU计算资源下限时，将优先接受计算任务，直到达到CPU计算资源上限；
其他处于上限和下限之间的情况，获取任务时可以进行随机分配。
步骤1205：现有集群中的主节点调用移动云计算中间平台接口，提交Hadoop任务进行运行。
步骤1206：集群中的数据节点完成任务计算，并将任务结果反馈给主节点。
本实施例中，假设某个数据节点如步骤1200中加入现有集群的智能终端申请退出集群，包括以下步骤：
步骤1207：该智能终端向集群中的主节点申请退出集群。
步骤1208：主节点进行数据资源迁移以及计算资源整合。实现如步骤1004所述，这里不再赘述。
步骤1209：资源整合完成后，同意该智能终端退出集群。
本实施例为网络接通后确定集群模型为定制机架结构的情况。
首先，在机架安装、上电，机架统一将交流电转化为直流电为所有设备提供电源后，进行版本镜像操作，即将图1所示整个系统打包作为镜像直接存储在ROM中。
安装完系统的硬件存储到备份件所在的机架中轴入口。初始安装因为需 要安装的设备数量较多，需要人为手工操作硬件供给。后续的备份设备以自动存储5-10块设备为佳。
通过与备份件替换相同的流程将硬件安装到指定的位置，如图13所示，图13为本发明实施例定制机架结构中机架的俯视图，如图13所示，计算存储模块安装于机架的圆环上。本实施例中，圆环以低速进行转动。
一方面，圆环的低速转动带来了传统设备的风冷散热的效果，另一方面，在硬件需要替换时，圆环的低速转动实现了计算存储备份模块通道的对接，以便于机架将替换的模块运送到位；同时，离心力能够自动拆除损坏的硬件。
然后，硬件联通机架交换设备自动获取网络参数，加入分布式集群。
最先合入集群的设备为主节点，可以在外部通过网络连接主节点的云计算任务调度平台执行分布式计算任务。
接着，加入备份设备，后续定时检测机架的备份设备数量，以便及时添加。同时需要回收损坏的硬件设备。
图14为本发明实施例定制机架结构中机架侧视图，如图14所示，机架的圆环槽位层可以是多层设计，带箭头的线表示了备份硬件的运行路径。
在进行硬件替换时，如图15所示，硬件槽位打开锁定，损坏硬件因离心力脱离机架；备份硬件经传送装置运行到备份导槽中；圆环逐渐降低速度至停止转动，使备份导槽和硬件需要替换的槽位对接，再经过传送装置的操作将备份硬件装入硬件槽位；更换备用硬件后，圆环逐渐回复原有运行旋转速度。
本实施例以典型应用场景对本发明实施例移动云计算中间平台及系统的工作实现进行描述。假设由于一游戏因计算量大且产生数据多，在一台手机上运行非常卡顿，同时需要基于这些大数据进行人工智能分析和挖掘，提供玩家场景发现和玩友推荐等，本实施例中假设在一个房间，有5个人，每人手持1台Android系统的智能手机，来测试使用这款大型游戏。图16为本发明实施例移动云计算中间平台应用的实施例的流程示意图，本实施例中，为例描述方便，仅以2台设备即Master主节点与一台Slave节点之间的业务流程为例，Master与其他Slave节点之间的业务流程与图16所示的业 务流程一样，这里不再赘述。如图16所示，包括：
步骤1600：分别在Master节点和Slave节点上安装移动云计算中间平台、Hadoop、游戏应用。
步骤1601：使用移动云计算中间平台对Android系统及Hadoop平台进行相关配置并生效，大致包括：
主节点的移动云计算中间平台为Hadoop平台提供Android系统缺少的Java库；主节点的移动云计算中间平台对Android系统进行硬件选取、加载；主节点的移动云计算中间平台分配一个IP地址，根据反馈的不同网络连接方式进行相应IP地址的配置；智能终端的Android系统读取分配的IP地址写入Hadoop配置文件并生效；主节点的移动云计算中间平台在Hadoop平台中进行集群内数据资源迁移以及计算资源整合，并通知集群内的其他Slave节点内进行资源调整。实现如图2所示，这里不再赘述。
步骤1602：一个或多个节点如图16中的Master节点和Slave节点检测游戏应用占用资源情况，并反馈到各自的Hadoop平台；一个或多个Slave节点的Hadoop平台收到游戏应用占用资源情况后，反馈到Master节点。
步骤1603：Master节点根据一个或多个节点存储和计算资源情况，自动进行数据资源迁移以及计算资源整合。
步骤1604：游戏应用开始使用，Master节点收到使用行为和任务后，为一个或多个Slave节点分配计算和存储任务；一个或多个Slave节点向Master节点反馈计算结果和存储结果，Master节点进行计算结果和存储结果汇总，并向游戏应用反馈游戏应用运算结果和存储结果。
步骤1605：在游戏应用的计算任务和存储任务进行过程中，主节点的移动云计算中间平台进行执行以下至少一种：
日志的存储，使用FIFO方式存储；硬件健康监控。
步骤1606：游戏应用结束，Master节点通知一个或多个Slave节点释放计算资源和存储资源；一个或多个Slave节点向Master节点反馈计算资源和存储资源释放结果。
本发明实施例还提供了一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被处理器执行时实现上述实施例所述的方法。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件、固件、硬件及其适当的组合。在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理单元的划分；例如，一个物理组件可以具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。某些组件或所有组件可以被实施为由处理器，如数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。
以上所述，仅为本发明的较佳实例而已，并非用于限定本发明的保护范围。凡在本发明的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。
工业实用性
本发明实施例移动云计算中间平台建立在Hadoop和Android系统之间，用于保障Hadoop和Android系统之间的适配以及运行，使得Hadoop实现了在Android系统上的正常运行即使得Android系统支持了Hadoop的运行。另一方面，通过本发明实施例移动云计算中间平台，智能终端实现了非常方便的组建、加入、移出无线网络或有线网络组建的分布式系统，特别地实现了使用更为廉价且方便更换的智能终端组建分布式集群。

Claims (19)

  1. 一种实现移动云计算中间平台的方法，包括：
    进行网络配置；根据加入分布式系统的智能终端设置Hadoop平台的运行配置(S1701)；
    对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理(S1702)；
    根据网络配置的信息及整合后的计算和存储资源进行云计算管理(S1703)；
    所述移动云计算中间平台建立在Hadoop平台和安卓Android系统之间。
  2. 根据权利要求1所述的方法，还包括
    为所述Hadoop平台提供所述Android系统缺少的Java库。
  3. 根据权利要求1所述的方法，其中，所述网络配置(S1701)包括：
    所述移动云计算中间平台第一次启动时，为安装所述移动云计算中间平台的智能终端分配一个固定地址；
    根据该智能终端所属分布式系统的连接方式获取互联网协议IP地址。
  4. 根据权利要求3所述的方法，其中，所述根据智能终端所属分布式系统的连接方式获取IP地址包括：
    所述连接方式为有线连接时，搜寻所述分布式系统的所有智能终端的IP地址，对最大的IP地址进行处理后作为所述IP地址；
    所述连接方式为无线连接时，所述IP地址由无线网动态主机配置协议DHCP自动分配；
    所述连接方式为定制机架方式时，根据槽位的不同来获取所述IP地址。
  5. 根据权利要求1所述的方法，其中，所述对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理(S1702)包括：
    当安装所述移动云计算中间平台的智能终端所属分布式系统接收到写入和计算任务时，确认当前分布式系统中的资源使用状况；
    根据确认的资源使用状况调取所述Hadoop平台的任务分配方法：当存储大小超出设定限额上限的不接受写入操作，当计算能力超出设定限额上限的不接受计算任务；当存储大小低于设定限额下限时，优先写入，直到达到设定限额的上限；当计算能力低于设定限额下限时，优先接受计算任务，直到达到设定限额的上限；所述分布式系统中存储大小和计算能力处于上限和下限之间的智能终端获取任务时进行随机分配。
  6. 根据权利要求1～5任一项所述的方法，还包括：
    为安装所述移动云计算中间平台的智能终端所述分布式系统中的智能终端提供硬件健康监控，并提供修复操作。
  7. 根据权利要求6所述的方法，还包括：
    当所述硬件出现问题后，进行重连或重启操作，预设次数重连或重启仍失败后，提示替换所述硬件。
  8. 根据权利要求1～5任一项所述的方法，还包括：
    当系统日志大于设定的范围值时，进行先入先出FIFO存储。
  9. 一种用于实现移动云计算中间平台的装置，至少包括存储器和处理器，其中，存储器中存储有以下可执行指令：
    进行网络配置；根据加入分布式系统的智能终端设置Hadoop平台的运行配置；对计算和存储的资源进行管理，基于计算和存储资源进行资源整合处理；根据网络配置的信息及整合后的计算和存储资源进行云计算管理；
    所述移动云计算中间平台建立在Hadoop平台和安卓Android系统之间。
  10. 一种实现分布式的方法，包括：
    获取并安装移动云计算中间平台和Hadoop平台到智能终端(S1801)；
    移动云计算中间平台在Hadoop平台中进行所述智能终端所属分布式系统内数据资源迁移以及计算资源整合(S1802)；
    移动云计算中间平台接收来自Hadoop平台反馈的所述分布式系统中的智能终端的存储和计算资源情况，并进行云计算管理(S1803)。
  11. 根据权利要求10所述的方法，还包括：
    当有新的智能终端接入所述分布式系统时，
    所述移动云计算中间平台根据新接入的智能终端的存储和计算资源情况，重新进行数据资源迁移以及计算资源整合。
  12. 根据权利要求11所述的方法，还包括：
    当有新的智能终端接入所述分布式系统时，新的智能终端连接至所述分布式系统所在网络，获取并安装所述移动云计算中间平台及Hadoop平台；
    新的智能终端使用安装的移动云计算中间平台，按照所述分布式系统所在网络的连接方式配置Hadoop系统网络。
  13. 根据权利要求12所述的方法，还包括：
    根据智能终端所属分布式系统的连接方式获取互联网协议IP地址，包括：
    所述连接方式为有线连接时，搜寻所述分布式系统的所有智能终端的IP地址，对最大的IP地址进行处理后作为所述IP地址；
    所述连接方式为无线连接时，所述IP地址由无线网动态主机配置协议DHCP自动分配；
    所述连接方式为定制机架方式时，根据槽位的不同来获取所述IP地址。
  14. 根据权利要求10所述的方法，还包括：
    所述获取并安装移动云计算中间平台和Hadoop平台之后，所述进行分布式系统内数据资源迁移以及计算资源整合之前，所述移动云计算中间平台提供所述智能终端的Android系统缺少的Java库，对所述智能终端的Android系统进行硬件选取、加载；
    所述移动云计算中间平台根据智能终端返回的分布式系统连接方式为所述智能终端分配IP地址；并写入Hadoop平台的配置文件。
  15. 根据权利要求10或11所述的方法，其中，所述进行数据资源迁移以及计算资源整合包括：
    存储大小超出设定限额上限的智能终端不接受写入操作，计算能力超出设定限额上限的职能终端不接受计算任务；
    当智能终端的存储大小低于设定限额下限时，优先写入，直到达到设定限额的上限；当智能终端的计算能力低于设定限额下限时，优先接受计算任务，直到达到设定限额的上限；
    存储大小和计算能力在上限和下限之间的智能终端获取任务时，进行随机分配。
  16. 根据权利要求10、11或12所述的方法，还包括：
    所述移动云计算中间平台采用先入先出FIFO存储方式存储日志。
  17. 根据权利要求10、11或12所述的集方法，还包括：
    所述移动云计算中间平台根据所述Hadoop平台反馈的所述分布式系统接入的智能终端的心跳情况进行硬件健康状况的监控。
  18. 根据权利要求10、11或12所述的方法，其中，所述移动云计算中间平台提供给外部接口进行云计算管理。
  19. 根据权利要求10、11或12所述的方法，还包括：
    当所述分布式系统中有智能终端退出所述分布式系统时，所述移动云计算中间平台接收到来自所述Hadoop平台的退出分布式系统申请，在所述Hadoop平台中再次进行所述分布式系统内数据资源迁移以及计算资源整合；
    在所述智能终端退出成功后，所述移动云计算中间平台记录日志。
PCT/CN2017/113647 2016-11-29 2017-11-29 实现移动云计算中间平台的方法及实现分布式的方法 WO2018099406A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611074053.3 2016-11-29
CN201611074053.3A CN106790403B (zh) 2016-11-29 2016-11-29 实现移动云计算中间平台的方法及实现分布式的方法

Publications (1)

Publication Number Publication Date
WO2018099406A1 true WO2018099406A1 (zh) 2018-06-07

Family

ID=58898589

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/113647 WO2018099406A1 (zh) 2016-11-29 2017-11-29 实现移动云计算中间平台的方法及实现分布式的方法

Country Status (2)

Country Link
CN (1) CN106790403B (zh)
WO (1) WO2018099406A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112905319A (zh) * 2021-02-04 2021-06-04 重庆惠科金渝光电科技有限公司 用于移动云服务位置调整的方法、装置及设备

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106790403B (zh) * 2016-11-29 2022-01-25 中兴通讯股份有限公司 实现移动云计算中间平台的方法及实现分布式的方法
CN109246479A (zh) * 2018-10-09 2019-01-18 深圳市亿联智能有限公司 一种基于智能机顶盒的云计算控制方式
CN113254505B (zh) * 2021-06-17 2021-10-08 湖南视觉伟业智能科技有限公司 分布式数据存储方法、检索方法、系统及可读存储介质
CN114866565B (zh) * 2022-04-20 2024-03-22 北京红山信息科技研究院有限公司 一种基于pass平台软件资源用分配系统

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105321124A (zh) * 2015-11-23 2016-02-10 南京信息工程大学 一种基于Hadoop的电力云平台设计方案
CN106790403A (zh) * 2016-11-29 2017-05-31 中兴通讯股份有限公司 实现移动云计算中间平台的方法及实现分布式的方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
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
US9213580B2 (en) * 2012-01-27 2015-12-15 MicroTechnologies LLC Transportable private cloud computing platform and associated method of use
US9268590B2 (en) * 2012-02-29 2016-02-23 Vmware, Inc. Provisioning a cluster of distributed computing platform based on placement strategy
CN103414761B (zh) * 2013-07-23 2017-02-08 北京工业大学 一种基于Hadoop架构的移动终端云资源调度方法
CN103399787B (zh) * 2013-08-06 2016-09-14 北京华胜天成科技股份有限公司 一种基于Hadoop云计算平台的MapReduce作业流式调度方法及调度系统
US9037646B2 (en) * 2013-10-08 2015-05-19 Alef Mobitech Inc. System and method of delivering data that provides service differentiation and monetization in mobile data networks
KR20150054505A (ko) * 2013-11-12 2015-05-20 건국대학교 산학협력단 홈 가전기기들에 대한 관리 서비스를 제공하는 클라우드 기반의 데이터 서버 및 홈 가전기기들에 대한 관리 서비스 제공 방법
CN103813213B (zh) * 2014-02-25 2019-02-12 南京工业大学 基于移动云计算的实时视频分享平台和方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105321124A (zh) * 2015-11-23 2016-02-10 南京信息工程大学 一种基于Hadoop的电力云平台设计方案
CN106790403A (zh) * 2016-11-29 2017-05-31 中兴通讯股份有限公司 实现移动云计算中间平台的方法及实现分布式的方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WANG HUIJUAN ET AL.: "Application of hadoop in mobile cloud computing", JOURNAL OF NORTH CHINA INSTITUTE OF AEROSPACE ENGINEERING, vol. 26, no. 3, 30 June 2016 (2016-06-30) *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112905319A (zh) * 2021-02-04 2021-06-04 重庆惠科金渝光电科技有限公司 用于移动云服务位置调整的方法、装置及设备
CN112905319B (zh) * 2021-02-04 2024-04-02 重庆惠科金渝光电科技有限公司 用于移动云服务位置调整的方法、装置及设备

Also Published As

Publication number Publication date
CN106790403B (zh) 2022-01-25
CN106790403A (zh) 2017-05-31

Similar Documents

Publication Publication Date Title
WO2018099406A1 (zh) 实现移动云计算中间平台的方法及实现分布式的方法
US10887247B2 (en) Dynamic resource allocation for sensor devices on a cellular network
CN109768871B (zh) 配置多个虚拟网卡的方法、宿主机和存储介质
KR101638436B1 (ko) 클라우드 스토리지 및 그의 관리 방법
US20180210853A1 (en) Extending the capabilities of existing devices without making modifications to the existing devices
US8499191B2 (en) Failure recovery method for information processing service and virtual machine image generation apparatus
JP5708937B2 (ja) 構成情報管理システム、構成情報管理方法、及び構成情報管理用プログラム
US8819190B2 (en) Management of file images in a virtual environment
CN103618627B (zh) 一种管理虚拟机的方法、装置及系统
US11809901B2 (en) Migrating the runtime state of a container between two nodes
WO2020248507A1 (zh) 基于容器云的系统资源监控方法及相关设备
WO2013177968A1 (zh) 文件存储系统、装置及文件存取方法
CN103530193A (zh) 用于调节应用进程的方法和设备
CN112162821B (zh) 容器集群资源监视方法、装置及系统
TW201741901A (zh) 資料遷移方法和裝置
US8996645B2 (en) Transmitting data by means of storage area network
US20120151095A1 (en) Enforcing logical unit (lu) persistent reservations upon a shared virtual storage device
TWI712876B (zh) 管理儲存子系統之電力消耗的電腦實現方法與電腦系統
CN105739930A (zh) 一种存储架构及其初始化方法和数据存储方法及管理装置
EP3018585A1 (en) Machine provision method, machine provision system, and machine provision program
US20150212902A1 (en) Network attached storage device with automatically configured distributed file system and fast access from local computer client
CN104636441A (zh) 网络文件系统实现方法和装置
JP2013003691A (ja) 計算機システムおよびその計算機システムにおけるディスク共有方法
CN115604294A (zh) 一种管理存储资源的方法及装置
CN107704618A (zh) 一种基于aufs文件系统的热迁徙方法和系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17876062

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17876062

Country of ref document: EP

Kind code of ref document: A1