CN103544064A - Cloud computing method, cloud management platform and client - Google Patents
Cloud computing method, cloud management platform and client Download PDFInfo
- Publication number
- CN103544064A CN103544064A CN201310517736.1A CN201310517736A CN103544064A CN 103544064 A CN103544064 A CN 103544064A CN 201310517736 A CN201310517736 A CN 201310517736A CN 103544064 A CN103544064 A CN 103544064A
- Authority
- CN
- China
- Prior art keywords
- calculation
- result
- management platform
- cloud management
- client
- 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
Images
Landscapes
- Information Transfer Between Computers (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
An embodiment of the invention provides a cloud computing method, a cloud management platform and a client. The method includes the steps: receiving a cloud resource request transmitted by user equipment by the cloud management platform; determining computing resources distributed for the user equipment from idle computing resources according to computing resource information in the cloud resource request by the cloud management platform; transmitting computing tasks to at least one client providing the computing resources by the cloud management platform, and accordingly, respectively fulfilling the computing tasks through running virtual machines by the clients. The cloud resource request contains the computing tasks and the computing resource information needed for fulfilling the computing tasks, the computing resources are provided by at least one client, and at least one virtual machine runs on each client. As the virtual machines run on the clients, functions of the virtual machines are as same as those of solid computers, and universal computing tasks can be fulfilled on the clients.
Description
Technical field
The embodiment of the present invention relates to data communication technology, relates in particular to a kind of cloud computing method, cloud management platform and client.
Background technology
Publicly-owned cloud service is the focus of current IT field, and the service provider of publicly-owned cloud generally has larger data center, and utilizes these data centers to provide service to medium-sized and small enterprises and individual.Common publicly-owned cloud operator operation virtualization software in the data center's infrastructure (server, the network equipment, memory device etc.) of oneself, after physical infrastructure is virtualized into virtual infrastructure, externally hire out again the virtual computational resources such as fictitious host computer, according to use amount charging.Medium-sized and small enterprises are used publicly-owned cloud service to obtain fast computing power, have avoided the disposable IT initial stage to drop into, and have possessed the ability with portfolio elastic telescopic.
At field of scientific study, propose aspiration and calculated (Volunteer Computing), for example BOINC.The scheme of BOINC provides utilizes the platform that on internet, the idle computing power of thousands of personal computers is carried out distributed science calculating.The developer of science computational item can utilize this platform, after the task burst of the huge computing power of needs, distributes to the computing machine that the participant that support to calculate has and moves.But, developer need to formulate different algorithms for different tasks, also to how consider by calculation task burst for different tasks, then, assignment algorithm is packaged in software package and is downloaded for volunteer, therefore, BOINC can not support general-purpose computations task, only supports specific distributed computing task.
Summary of the invention
The embodiment of the present invention provides a kind of cloud computing method, cloud management platform and client, can support general-purpose computations task.
First aspect present invention provides a kind of cloud computing method, comprising:
Cloud management platform receives the cloud resource request that subscriber equipment sends, and comprises calculation task and complete the required computational resource information of described calculation task in described cloud resource request;
Described cloud management platform is according to the described computational resource information comprising in described cloud resource request, from idle computational resource, determine the computational resource into described user equipment allocation, described computational resource is provided by least one client, and in each client, operation has at least one virtual machine;
Described cloud management platform described calculation task is provided provide at least one of described computational resource client, make described each client by described each virtual machine of operation, complete described calculation task respectively, and return to described cloud management platform the result of calculation that described calculation task is corresponding;
Described cloud management platform receives the result of calculation that described calculation task is corresponding, according to described result of calculation, determines the final calculation result that described calculation task is corresponding.
In the possible implementation of the first of first aspect present invention, described cloud management platform, according to described computational resource information, is determined the computational resource into described user equipment allocation from idle computational resource, comprising:
Described cloud management platform is according to described computational resource information, and described subscriber equipment and the Internet resources between each client of described idle computational resource are provided, and determines the described computational resource into described user equipment allocation from described idling-resource.
In the possible implementation of the second of first aspect present invention, described cloud management platform receives described result of calculation, after determining final calculation result, also comprises according to described result of calculation:
Described cloud management platform receives the resource release request that described subscriber equipment sends;
Described cloud management platform discharges according to described resource release request the described computational resource that described at least one client provides.
In the third possible implementation of first aspect present invention, also comprise:
Described cloud management platform receives the idle computational resource information that described each client reports.
In the 4th kind of possible implementation of first aspect present invention, described idle computational resource information at least comprises: the surplus capacity of the central processing unit CPU of described each client, free memory space, remanence disk space.
Four kinds of possible implementations of the first to the in conjunction with first aspect present invention and first aspect, in the 5th kind of possible implementation of first aspect present invention, the computing unit tolerance of described computational resource after with standardization, each computing unit has identical CPU processing power and memory headroom, disk space, in described cloud resource request, comprise to complete the required computational resource information of described calculation task be the number N of the required computing unit of described calculation task;
Described cloud management platform, according to described computational resource information, is determined the computational resource into described user equipment allocation from idle computational resource, comprising:
Described cloud management platform, according to the number N that completes the required computing unit of described calculation task, is determined N the computing unit into described user equipment allocation from idle computing unit.
In the 6th kind of possible implementation of first aspect present invention, in described cloud resource request, also comprise:
Redundant resource sign, described redundant resource sign is used for to described cloud management platform application three batch total operator resources, and described every batch total operator resource comprises respectively N computing unit, and described every batch total operator resource has been respectively used to described calculation task;
Accordingly, described cloud management platform, according to described computational resource information, is determined as described user equipment allocation computational resource from idle computational resource, comprising:
Described cloud management platform, according to described redundant computation sign and described computational resource information, is determined the three batch total operator resources into described user equipment allocation from described idle computing unit;
Described cloud management platform described calculation task is provided provide at least one of described computational resource client, make described each client by described each virtual machine of operation, complete described calculation task respectively, and return to described cloud management platform the result of calculation that described calculation task is corresponding, comprising:
Described cloud management platform sends to described calculation task respectively each client that described three batch total operator resources are provided, so that each virtual machine moving in described each client calculates respectively described calculation task, and return to described cloud management platform three result of calculations that described calculation task is corresponding;
Described cloud management platform receives the result of calculation that described calculation task is corresponding, according to described result of calculation, determines the final calculation result that described calculation task is corresponding, comprising:
Described cloud management platform receives three result of calculations that described calculation task is corresponding, according to three result of calculations that described calculation task is corresponding, determines the final calculation result that described calculation task is corresponding.
In the 7th kind of possible implementation of first aspect present invention, if described calculation task needs M step just can complete calculating, a checkpoint is set after each step, described checkpoint is for when described each client moves to described checkpoint, and described each client reports to described cloud management platform by the result of calculation of this step separately respectively;
Correspondingly, described cloud management platform receives three result of calculations that described calculation task is corresponding, according to three result of calculations that described calculation task is corresponding, determines the final calculation result that described calculation task is corresponding, comprising:
Described cloud management platform, according to the result of calculation that formerly receives this step separately that the client in two batch total operator resources reports respectively, judges that whether the result of calculation of this step that described two batch total operator resources report is identical;
If, the identical result of calculation that described cloud management platform reports described two batch total operator resources is as the final calculation result of this step, if described management platform is not also received the result of calculation of this step that another batch total operator resource reports, stop the operation of described another batch total operator resource;
Whether if not, described cloud management platform is waited for the result of calculation of this step that described another batch total operator resource reports, judge in the result of calculation of this step that described three batch total operator resources report and have two batch totals to come to the same thing;
If have two batch totals to come to the same thing in the result of calculation of this step that described three batch total operator resources report, the identical result of calculation that described cloud management platform reports described two batch total operator resources is as the final calculation result of this step;
Described cloud management platform is at synchronization, using the final calculation result of described step as next step initial value, send to respectively each client of described three batch total operator resources, so that starting next step according to the final calculation result of described step simultaneously, each client of described three batch total operator resources calculates, until M step finishes;
Described cloud management platform is using the net result of described M step as final calculation result corresponding to described calculation task.
In the 8th kind of possible implementation of first aspect present invention, if three result of calculations that in certain step, three batch total operator resources report are all different, described management platform finishes the calculation task of described three batch total operator resources.
Second aspect present invention provides a kind of cloud computing method, comprising:
Client creates at least one virtual machine in this locality, described each virtual machine is for providing computational resource to cloud management platform;
The calculation task that described in described client, cloud management platform sends, by moving described each virtual machine, utilize described computational resource to complete described calculation task, and result of calculation corresponding to described calculation task is sent to described cloud management platform, so that described cloud management platform is determined the final calculation result of described calculation task according to result of calculation corresponding to described calculation task.
In the possible implementation of the first of second aspect present invention, described client creates at least one virtual machine in this locality, comprising:
Described client is downloaded cloud service plug-in unit from described cloud management platform, at cloud service plug-in unit described in local runtime, creates described at least one virtual machine by described cloud service plug-in unit in this locality.
In the possible implementation of the second of second aspect present invention, also comprise:
Described client is obtained utilization factor, disk utilization, the memory usage of local central processing unit CPU, determine local idle computational resource, to described cloud management platform, report described idle computational resource, idle computational resource information at least comprises: the surplus capacity of the CPU of described each client, free memory space, remanence disk space.
In the third possible implementation of second aspect present invention, the calculation task that cloud management platform sends described in described client, utilizes after described computational resource completes described calculation task by moving described virtual machine, also comprises:
The resource release request that described in described client, cloud management platform sends, discharges described computational resource according to described resource release request.
In conjunction with the first of second aspect present invention and second aspect to the third possible implementation, in the 4th kind of possible implementation of second aspect present invention, the computing unit tolerance of described computational resource after with standardization, each computing unit position has identical CPU processing power, memory headroom and disk space;
Correspondingly, described client reports described idle computational resource to described cloud management platform, comprising:
Described client reports the number of idle computing unit to described cloud management platform.
In the 5th kind of possible implementation of second aspect present invention, if described calculation task needs M step just can complete calculating, a checkpoint is set after each step;
When described client moves to described checkpoint, the result of calculation of this step is reported to described cloud management platform, so that described cloud management platform is according to the result of calculation of described step, determine the final calculation result of this step, using the final calculation result of this step as next step initial value, send to described client;
Accordingly, next step the initial value that described client also receives that described cloud management platform sends.
Third aspect present invention provides a kind of cloud management platform, comprising:
Receiver module, the cloud resource request sending for receiving subscriber equipment, comprises calculation task in described cloud resource request and completes the required computational resource information of described calculation task;
Resource determination module, for the described computational resource information comprising according to described cloud resource request, from idle computational resource, determine the computational resource into described user equipment allocation, described computational resource is provided by least one client, and in each client, operation has at least one virtual machine;
Sending module, for described calculation task is sent at least one of described computational resource client is provided, make described each client by described each virtual machine of operation, complete described calculation task respectively, and return to described cloud management platform the result of calculation that described calculation task is corresponding;
Described receiver module, also for receiving the result of calculation that described calculation task is corresponding;
Result of calculation determination module, determines for the described result of calculation receiving according to described receiver module the final calculation result that described calculation task is corresponding.
In the possible implementation of the first of third aspect present invention, described resource determination module specifically for:
According to described computational resource information, and described subscriber equipment and the Internet resources between each client of described idle computational resource are provided, from described idling-resource, determine the described computational resource into described user equipment allocation.
In the possible implementation of the second of third aspect present invention, also comprise: resource release module;
Described receiver module also for: receive the resource release request that described subscriber equipment sends;
Described resource release module, for discharging the described computational resource that described at least one client provides according to described resource release request.
In the third possible implementation of third aspect present invention, described receiver module also for:
Receive the idle computational resource information that described each client reports.
In the 4th kind of possible implementation of third aspect present invention, described idle computational resource information at least comprises: the surplus capacity of the central processing unit CPU of described each client, free memory space, remanence disk space.
Four kinds of possible implementations of the first to the in conjunction with third aspect present invention and the third aspect, in the 5th kind of possible implementation of third aspect present invention, the computing unit tolerance of described computational resource after with standardization, each computing unit has identical CPU processing power and memory headroom, disk space, in described cloud resource request, comprise to complete the required computational resource information of described calculation task be the number N of the required computing unit of described calculation task;
Described resource determination module specifically for: according to the number N that completes the required computing unit of described calculation task, from idle computing unit, determine N the computing unit into described user equipment allocation.
In the 6th kind of possible implementation of third aspect present invention, in described cloud resource request, also comprise:
Redundant resource sign, described redundant resource sign is used for to described cloud management platform application three batch total operator resources, and described every batch total operator resource comprises respectively N computing unit, and described every batch total operator resource has been respectively used to described calculation task;
Accordingly, described resource determination module specifically for: according to described redundant computation sign and described computational resource information, from described idle computing unit, determine the three batch total operator resources into described user equipment allocation;
Described sending module specifically for: described calculation task is sent to respectively to each client that described three batch total operator resources are provided, so that each virtual machine moving in described each client calculates respectively described calculation task, and return to described cloud management platform three result of calculations that described calculation task is corresponding;
Described receiver module specifically for: receive three result of calculations that described calculation task is corresponding;
Described result of calculation determination module specifically for: according to three result of calculations that described calculation task is corresponding, determine the final calculation result that described calculation task is corresponding.
In the 7th kind of possible implementation of third aspect present invention, if described calculation task needs M step just can complete calculating, a checkpoint is set after each step, described checkpoint is for when described each client moves to described checkpoint, and described each client reports to described cloud management platform by the result of calculation of this step separately respectively;
Correspondingly, result of calculation determination module specifically for: the result of calculation of this step separately reporting respectively according to the client formerly receiving in two batch total operator resources, judges that whether the result of calculation of this step that described two batch total operator resources report identical;
If, the identical result of calculation described two batch total operator resources being reported is as the final calculation result of this step, if described management platform is not also received the result of calculation of this step that another batch total operator resource reports, stop the operation of described another batch total operator resource;
If not, wait for the result of calculation of this step that described another batch total operator resource reports, judge in the result of calculation of this step that described three batch total operator resources report and whether have two batch totals to come to the same thing;
If have two batch totals to come to the same thing in the result of calculation of this step that described three batch total operator resources report, the identical result of calculation described two batch total operator resources being reported is as the final calculation result of this step;
At synchronization, using the final calculation result of described step as next step initial value, send to respectively each client of described three batch total operator resources, so that starting next step according to the final calculation result of described step simultaneously, each client of described three batch total operator resources calculates, until M step finishes;
Using the net result of described M step as final calculation result corresponding to described calculation task.
In the 8th kind of possible implementation of third aspect present invention, if three result of calculations that in certain step, three batch total operator resources report are all different, described result of calculation determination module is specifically for the calculation task that finishes described three batch total operator resources.
Fourth aspect present invention provides a kind of client, comprising:
Virtual machine creating module, for create at least virtual machine in this locality, described each virtual machine is for providing computational resource to cloud management platform;
Receiver module, the calculation task sending for receiving described cloud management platform;
Computing module, for utilizing described computational resource to complete described calculation task by moving described each virtual machine;
Sending module, for result of calculation corresponding to described calculation task is sent to described cloud management platform, so that described cloud management platform is determined the final calculation result of described calculation task according to result of calculation corresponding to described calculation task.
In the possible implementation of the first of fourth aspect present invention, described virtual machine creating module specifically for:
From described cloud management platform, download cloud service plug-in unit, at cloud service plug-in unit described in local runtime, by described cloud service plug-in unit, in this locality, create described at least one virtual machine.
In the possible implementation of the second of fourth aspect present invention, also comprise:
Idling-resource reporting module, for obtaining utilization factor, disk utilization, the memory usage of local central processing unit CPU, determine local idle computational resource, to described cloud management platform, report described idle computational resource, idle computational resource information at least comprises: the surplus capacity of the CPU of described each client, free memory space, remanence disk space.
In the third possible implementation of fourth aspect present invention, also comprise: resource release module;
Described receiver module also for: receive the resource release request that described cloud management platform sends;
Described resource release module, for discharging described computational resource according to described resource release request.
In conjunction with the first of fourth aspect present invention and fourth aspect to the third possible implementation, in the 4th kind of possible implementation of fourth aspect present invention, the computing unit tolerance of described computational resource after with standardization, each computing unit position has identical CPU processing power, memory headroom and disk space;
Correspondingly, described idling-resource reporting module is specifically for number from idle computing unit to described cloud management platform that report.
In the 5th kind of possible implementation of fourth aspect present invention, if described calculation task needs M step just can complete calculating, a checkpoint is set after each step;
Described sending module specifically for: when moving to described checkpoint, the result of calculation of this step is reported to described cloud management platform, so that described cloud management platform is according to the result of calculation of described step, determine the final calculation result of this step, using the final calculation result of this step as next step initial value, send to described client;
Accordingly, described receiver module is also for, the initial value of next step that receives that described cloud management platform sends.
The cloud computing method of the embodiment of the present invention, cloud management platform and client, the cloud resource request that cloud management platform sends by subscriber equipment, be defined as the computational resource of subscriber equipment for distributing, and calculation task is sent at least one of computational resource client is provided, make each client by each virtual machine of operation, complete calculation task respectively, and return to cloud management platform the result of calculation that calculation task is corresponding.Because operation in client has virtual machine, the function of virtual machine is identical with entity computer, therefore in client, can realize general-purpose computations task.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the process flow diagram of cloud computing method embodiment mono-of the present invention;
Fig. 2 is the process flow diagram of cloud computing method embodiment bis-of the present invention;
Fig. 3 is the process flow diagram of cloud computing method embodiment tri-of the present invention;
Fig. 4 is the applicable redundant computation process flow diagram of the embodiment of the present invention;
Fig. 5 is the process flow diagram of cloud computing method embodiment tetra-of the present invention;
Fig. 6 is the structural representation of cloud management platform embodiment mono-of the present invention;
Fig. 7 is the structural representation of cloud management platform embodiment bis-of the present invention;
Fig. 8 is the structural representation of client implementation example one of the present invention;
Fig. 9 is the structural representation of cloud management platform embodiment tri-of the present invention;
Figure 10 is the structural representation of client implementation example two of the present invention.
Embodiment
For making object, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.
Before introducing technical scheme provided by the invention, first simply introduce the function of the network equipment used in various embodiments of the present invention, cloud management platform is carried out maintenance and management by cloud service provider (Ye Cheng operator once), operator (Operator) sets up a cloud computing service based on idle local computing ability, to all users, provides IaaS cloud computing service.Operator's operation cloud management platform (platform), allow supplier (Provider) registration local computing ability, and provide application program cloud service plug-in unit (cloud plug-in) for supplier's downloading-running, supplier refers to and wish to utilize individual that idle computational resource acquires an advantage or the client of enterprise, supplier has the local computer (being client) of oneself, when supplier is after cloud service platform registering account, need to cloud service plug-in unit be installed in client downloads, client can connect with cloud service platform by cloud service plug-in unit, and by the resource virtualizing in client, depending on CPU (central processing unit) (Central Processing Unit, be called for short CPU) and the situation that takies of internal memory, computational resource can be standardized as to computing unit (Compute Unit, be called for short CU).Cloud service plug-in unit is after client background operation, can in cloud management platform, register local available computing unit (free CU), a client is according to computing power situation and supplier's wish, the virtual one or more computing units of possibility, each computing unit and supplier's account number binding.User (User) can rent virtual computational resource in this cloud management platform, carries out the business such as data analysis, data mining, video code conversion, science calculating, business event logical process.
Fig. 1 is the process flow diagram of cloud computing method embodiment mono-of the present invention, and as shown in Figure 1, the method for the present embodiment can comprise:
When user has business demand, by subscriber equipment, sign in to cloud management platform, to cloud management platform, initiate cloud resource request, in cloud resource request, comprise calculation task and complete the required computational resource information of calculation task.Computational resource information can be size of the size of the internal memory that completes the required CPU processing power of calculation task, take, required disk space etc.In a kind of implementation of the present invention, by the virtual computing unit that turns to standard of computational resource, each computing unit has certain C PU processing power and memory headroom, disk space, and each computing unit size is identical.Therefore the required computational resource information of calculation task that completes, comprising in cloud resource request can be for completing the number N of the required computing unit of calculation task.
Step 102, cloud management platform are according to the computational resource information comprising in cloud resource request, from idle computational resource, determine the computational resource into user equipment allocation, computational resource is provided by least one client, and in each client, operation has at least one virtual machine.
Cloud management platform is after receiving cloud resource request, idle computational resource in scanning system, according to required computational resource information, from idle computational resource, determine the computational resource into user equipment allocation, for the computational resource of user assignment can be provided by a plurality of clients, each client completes a part of calculation task, and in each client, operation has at least one virtual machine.Idle computational resource information at least comprises: the surplus capacity of the CPU of each client, free memory space, remanence disk space.Each client under cloud management platform can active reporting oneself idle computational resource.
Cloud management platform is according to computational resource information, from idle computational resource, determine the computational resource into user equipment allocation, be specially: cloud management platform is according to computational resource information, and subscriber equipment and the Internet resources between each client of idle computational resource are provided, from idling-resource, determine the computational resource into user equipment allocation.If computational resource is measured with computing unit, cloud management platform, according to the network condition between subscriber equipment and the client at each computing unit place, as the bandwidth between subscriber equipment and computing unit, is chosen one group of computing unit of full blast.In the resource request of the initiation of subscriber equipment, IP address and the network condition that can carry subscriber equipment, so that cloud management platform is chosen one group of computing unit of full blast according to the IP address of subscriber equipment and network condition.
Client is by downloading cloud service plug-in unit from cloud management platform, in the tangible client of cloud service plug-in unit, realize the key of general-purpose computations, client is set up virtual machine by cloud service plug-in unit, virtual adding, is equivalent to a computing machine, and this virtual machine can be realized general computing power as real computing machine.
To simply introduce the major function of cloud service plug-in unit below:
(1) for secure communication layer, provide the communication link of encryption, for example adopt security socket layer (Secure Sockets Layer, being called for short SSL) cryptographic algorithm encrypts communication link, and this communication link is not only as the link of service request, also as the communication link of the virtual machine in client.Client is communicated by letter with cloud management platform, accept resource granting and resource release request from cloud management platform.
(2) complete scheduling and the measurement of local resource, by measurement, determine whether to have enough resources to accept the request from cloud management platform, and measurement result is regularly reported to cloud management platform.The project of measuring comprises the occupancy of CPU, internal memory and disk, local idle computing power assessment etc.
(3) data encryption in the virtual machine of user being rented, mainly comprises the encryption of internal memory, hard disk, network, and internal memory is encrypted: the data of user in virtual memory need to be stored in the physical memory of client after encrypting again; HD encryption: this locality storage data in user's virtual machine need to be stored in the local storage medium of client after encrypting again; Network encryption: user's virtual machine and extraneous all IP packets needs after encrypting again the physical network card by client transmit.By encryption technology, thereby can guarantee the security of user's data, make the supplier of this client and other staff all cannot see the user data of storing on virtual machine.
(4) virtualization layer, by set up operation virtual machine on the operating system of client, provides operation mainstream user the ability of operating system, and by the virtual computing unit that turns to standard of local cpu, internal memory and hard disk.
In cloud management platform, determine after the computational resource of distributing to user, cloud management platform can communicate with the client of calculating is provided, cloud management platform is communicated by letter with the cloud service plug-in unit moving in selected client, on the virtual machine that distribution of computation tasks is moved in this client, virtual machine is carried out calculation task, and returns to cloud management platform the result of calculation that calculation task is corresponding.
The method that the present embodiment provides, the cloud resource request that cloud management platform sends by subscriber equipment, be defined as the computational resource of subscriber equipment for distributing, and calculation task is sent at least one of computational resource client is provided, make each client by each virtual machine of operation, complete calculation task respectively, and return to cloud management platform the result of calculation that calculation task is corresponding.Because operation in client has virtual machine, the function of virtual machine is identical with entity computer, therefore in client, can realize general-purpose computations task.
Very huge owing to being connected at present the personal computer number of internet.Aspect computing power, due to popularizing of Intel's multi-core CPU, in fact personal user has been difficult to use up the CPU ability on own personal computer (personal computer is called for short PC).Network bandwidth aspect, individual and domestic consumer's access bandwidth has developed into even various data user's circuits (Digital Subscriber Line of 100M of 20M from 56K dialing in recent years, be called for short DSL) and light to family, access, the broadband charging mode of main flow is not limit flow of monthly payment.The average utilization of a large amount of personal computers is quite low, and the average utilization of same bandwidth is also very low.The method that the present embodiment provides, by utilizing the idle personal computer of individual and enterprise customer (client) and broadband access resource, set up a cloud computing service based on subscriber's local computing power, avoid operator in earlier stage to set up the huge input of data center, greatly reduce and assess the cost and entry threshold, set up a kind of new publicly-owned cloud business model and the ecosystem.Stand in supplier's angle, if these idle calculating and Internet resources can be sold by certain mode, at least offset the spending of part broadband, do not affect normal use simultaneously, be very attractive concerning individual and domestic consumer.Except individual and domestic consumer, a lot of enterprises also have a large amount of computational resources and bandwidth.It for main business, not the enterprise of ISP or cloud service, if can be idle computational resource and bandwidth usage in the situation that not affecting existing business, at least offset part IT expense, for enterprise, say it is also very valuablely by the present embodiment, by the idle computing power personal computer, to use, by a concentrated managed operation platform, publicly-owned cloud service is provided, thereby has met the demand of following three aspects:: 1, individual and enterprise wish to acquire an advantage by idle computational resource.2, IT enterprises wish to carry out publicly-owned cloud service with low cost, avoid a large amount of inputs at initial stage.3, enterprise wishes to obtain large capacity with low cost, the elasticity computing power of paying by use amount.
In the present embodiment, cloud management platform receives result of calculation, after determining final calculation result according to result of calculation, be after client has completed calculation task, cloud management platform also receives the resource release request that subscriber equipment sends, and then according to resource release request, discharges the computational resource that at least one client provides.
Fig. 2 is the process flow diagram of cloud computing method embodiment bis-of the present invention, because the individual PC resource providing is insecure, it is much higher that the likelihood ratio that break down or network disconnection etc. causes service to use plays self-built data center of operator.How in insecure resource, to realize reliably and calculating, be the key issue that the present embodiment will solve.In the present embodiment, by redundant computation, guarantee the reliability of result of calculation, below will illustrate.As shown in Figure 2, the method for the present embodiment can comprise:
In the present embodiment, also comprise redundant resource sign in cloud resource request, redundant resource sign is used for to cloud management platform application three batch total operator resources, and every batch total operator resource comprises respectively N computing unit, and every batch total operator resource has been respectively used to calculation task.
Every batch total operator resource comprises N computing unit, every group of computing unit can complete calculation task, cloud management platform sends to calculation task respectively each client of three batch total operator resources, so that client corresponding to each batch total operator resource calculated calculation task simultaneously, obtain three result of calculations, and return to cloud management platform three result of calculations that calculation task is corresponding.
In a kind of implementation, the cloud management platform result of calculation that relatively each batch total operator resource returns, if there are two result of calculations of a result of calculation and other different, think that this result of calculation makes mistakes, using other two identical result of calculations as final calculation result corresponding to calculation task.
In a kind of feasible implementation, if calculation task needs M step just can complete calculating, a checkpoint is set after each step, and checkpoint is for when each client moves to checkpoint, and each client reports to cloud management platform by the result of calculation of this step separately respectively.
Correspondingly, cloud management platform receives three result of calculations corresponding to calculation task, according to three result of calculations, determines the final calculation result that calculation task is corresponding, is specially:
Cloud management platform is according to the result of calculation that formerly receives this step separately that client corresponding to two batch total operator resources report respectively, whether the result of calculation that judges this step that two batch total operator resources report is identical, here, the client that every batch total operator resource is corresponding is all moved identical computational logic, after executing first step, each client reports respectively the result of calculation of this step, during the result of calculation of cloud management platform this step that client corresponding to two batch total operator resources reports before receiving, for example cloud management platform constantly receives first the result of calculation that client that the first computational resource is corresponding reports, second, constantly receive the result of calculation that client that the second computational resource is corresponding reports, wherein second be greater than first constantly constantly, cloud management platform constantly receives after the result of calculation that client that the second computational resource is corresponding reports second, relatively whether two result of calculations are identical.
If, the identical result of calculation that two batch total operator resources report is identical, the identical result of calculation that cloud management platform reports two batch total operator resources is as the final calculation result of this step, if management platform is not also received the result of calculation of this step that another batch total operator resource reports, stop the operation of another batch total operator resource, directly the final calculation result using identical result of calculation as this step.
If not, the identical result of calculation that two batch total operator resources report is not identical, cloud management platform is waited for the result of calculation of this step that another batch total operator resource reports, whether judge in the result of calculation of this step that three batch total operator resources report has two batch totals to come to the same thing, compare between two, determine whether that the result of calculation that two batch total operator resources report is identical, if have two batch totals to come to the same thing in the result of calculation of this step that three batch total operator resources report, the identical result of calculation that cloud management platform reports two batch total operator resources is as the final calculation result of this step,
In cloud management platform, determine after the result of calculation of this step, at synchronization, the final calculation result of this step is sent to respectively to each client of three batch total operator resources, as next step input parameter, so that each client of three batch total operator resources starts next step calculating according to the final calculation result of this step simultaneously, until M step finishes, cloud management platform is all carried out above-mentioned judgement in each step, and the input parameter using the technical result of previous step as next step, until finish in steps, cloud management platform is using the net result of M step as final calculation result corresponding to calculation task.
If three result of calculations that in certain step, three batch total operator resources report are all different, illustrate to calculate and made mistakes, management platform finishes the calculation task of three batch total operator resources, can restart to calculate.
The method that the present embodiment provides, by applying for other two groups of unnecessary computational resources, carry out calculation task simultaneously, client is after each step executes, all to cloud management platform, report the result of calculation of this step, the result of calculation of each step that cloud management platform is returned according to each batch total operator resource, determine the final calculation result of this step, discard wrong result of calculation, and the input parameter using the final calculation result of this step as next step, thereby guarantee that the result of calculation of each step is all accurately.Therefore, can guarantee the reliability of the final calculation result that calculation task is corresponding.
Fig. 3 is the process flow diagram of cloud computing method embodiment tri-of the present invention, the present embodiment will be elaborated to embodiment of the method shown in Fig. 2 by a concrete example, because the expense of each computing unit is more much lower than traditional publicly-owned cloud service, user can apply for using a plurality of computing units to process a mission critical simultaneously.For example apply for three virtual machines (Virtual Manufacturing is called for short VM)), move identical business software.As shown in Figure 4, Fig. 4 is the applicable redundant computation process flow diagram of the embodiment of the present invention, vm1 in Fig. 4, and vm2, vm3 represents respectively three virtual machines.Please refer to shown in Fig. 3 and Fig. 4, the method for the present embodiment can comprise:
Step 301, at vm1, vm2 inputs D1 simultaneously on vm3, start step1 calculation task.
Suppose that input D1 finally need to obtain exporting D4, computation process need to be through three steps (step).User arranges a checkpoint in business software after each step.Each run is to checkpoint, and virtual machine all can report cloud management platform result, and cloud management platform is as an arbitration platform.
The service logic of arbitration platform is: tentation data d1 need to obtain data d2 through processing, common calculating need to be used a computing unit, in order to improve reliability, adopt redundant computation, at vm1, vm2, on vm3, carry out same computational logic simultaneously, may constantly obtain d2 (vm1) at t1, may constantly obtain d2 (vm2) at t2, may constantly obtain d2 (vm3) at t3, suppose t1<t2<t3, the successful standard of redundant computation is: if d2 (vm1)=d2 (vm2), export d2 (vm1), and d2 (vm3) is forced to be set as d2(vm1), d2 (vm1) unequal to d2 (vm2) & & d2 (vm1)=d2 (vm3), export d2 (vm1), and d2 (vm2) is forced to be set as d2(vm1, d2 (vm1) unequal to d2 (vm2) & & d2 (vm2)=d2 (vm3), export d2 (vm2), and d2 (vm1) is forced to be set as d2(vm2), other situation redundant computation failure, need to recalculate.
In step1 checkpoint, cloud management platform obtains the result of calculation of vm1 and vm2 prior to vm3, the result of calculation of vm1 is d2 (vm1), the result of calculation of vm2 is d2 (vm2), then, cloud management platform judges that whether d2 (vm1) is identical with d2 (vm2), if the result of calculation of vm1 and vm2 is identical, cloud management platform stops vm3 and continues to calculate, and the net result of step1 is set as to d2 (vm1).
Three virtual machines are all used same d2 (vm1), start step2 calculate at same time point.
In step1 checkpoint, cloud management platform obtains the result of calculation of vm1 and vm3 prior to vm2, the result of calculation of vm1 is d3 (vm1), the result of calculation of vm3 is d3 (vm3), then, cloud management platform judges that whether d3 (vm1) is identical with d3 (vm3), if d3 (vm1) is identical with d3 (vm3), cloud management platform stops vm2 continuation calculating, and the net result of step2 is set as to d3 (vm1).
Three virtual machines are all used same d3 (vm1), at same time point, start next step calculating.
In step3 checkpoint, cloud management platform obtains the result of calculation of vm1 and vm2 prior to vm3, the result of calculation of vm1 is d4 (vm1), the result of calculation of vm2 is d4 (vm2), then, cloud management platform judges that whether d4 (vm1) is identical with d4 (vm2), if the result of calculation of vm1 and vm2 is identical, cloud management platform stops vm3 and continues to calculate, and the net result of step3 is set as to d4 (vm1).
Arbitration described above, result force to set, and starting next step calculating etc. is all the service that cloud management platform provides, and does not need user to realize voluntarily.Cloud management platform only need provide simple api interface to call for user program or script as arbitration platform.These api interfaces comprise: the result of 1, reporting certain step; 2, force to stop calculation procedure; 3, with certain input parameter, start calculation procedure.If user need to use redundant computation, the operation needing: 1, apply for a plurality of computational resources; 2, on each computational resource, move identical virtual machine image; 3, preset application program in virtual machine image realizes these api interfaces in application program.User does not need to consider arbitration and the scheduling process of Distributed Calculation, has greatly simplified the difficulty of using redundant computation.
It should be noted that, arbitration platform can be realized in cloud management platform, also can use the distributed algorithm based on election to realize on the cloud service plug-in unit of client.
Fig. 5 is the process flow diagram of cloud computing method embodiment tetra-of the present invention, and as shown in Figure 5, the cloud computing method that the present embodiment provides comprises:
Client creates at least virtual machine in this locality, be specially: client is downloaded cloud service plug-in unit from cloud management platform, at local runtime cloud service plug-in unit, creates at least one virtual machine by cloud service plug-in unit in this locality.Cloud service plug-in unit is the key that realizes general-purpose computations in client, and client is set up virtual machine by cloud service plug-in unit, and virtual adding is equivalent to a computing machine, and this virtual machine can be realized general computing power as real computing machine.To simply introduce the major function of cloud service plug-in unit below:
(1) for secure communication layer, provide the communication link of encryption, for example adopt security socket layer (Secure Sockets Layer, being called for short SSL) cryptographic algorithm encrypts communication link, and this communication link is not only as the link of service request, also as the communication link of the virtual machine in client.Client is communicated by letter with cloud management platform, accept resource granting and resource release request from cloud management platform.
(2) complete scheduling and the measurement of local resource, by measurement, determine whether to have enough resources to accept the request from cloud management platform, and measurement result is regularly reported to cloud management platform.The project of measuring comprises the occupancy of CPU, internal memory and disk, local idle computing power assessment etc.
(3) data encryption in the virtual machine of user being rented, mainly comprises the encryption of internal memory, hard disk, network, and internal memory is encrypted: the data of user in virtual memory need to be stored in the physical memory of client after encrypting again; HD encryption: this locality storage data in user's virtual machine need to be stored in the local storage medium of client after encrypting again; Network encryption: user's virtual machine and extraneous all IP packets needs after encrypting again the physical network card by client transmit.By encryption technology, thereby can guarantee the security of user's data, make the supplier of this client and other staff all cannot see the user data of storing on virtual machine.
(4) virtualization layer, by set up operation virtual machine on the operating system of client, provides operation mainstream user the ability of operating system, and by the virtual computing unit that turns to standard of local cpu, internal memory and hard disk.
The calculation task that step 402, client cloud management platform send, by moving each virtual machine, utilize computational resource to complete calculation task, and result of calculation corresponding to calculation task is sent to cloud management platform, so that cloud management platform is determined the final calculation result of calculation task according to result of calculation corresponding to calculation task.
Wherein, computational resource can be with the computing unit tolerance after standardization, each computing unit position has identical CPU processing power and memory headroom, disk space, therefore, the client that the present embodiment provides also has the virtualized function of local resource, client is by obtaining utilization factor, disk utilization, the memory usage of local cpu, determine local idle computational resource, then turn to after standardized computing unit the idle computational resource in this locality is virtual, to cloud management platform, report idle computational resource.If it is virtual that client is not carried out idle computational resource, idle computational resource information at least comprises: the surplus capacity of the CPU of each client, free memory space, remanence disk space.If it is virtual that client is carried out idle computational resource, the number that the idle computational resource reporting is idle computing unit.
The method that the present embodiment provides, client, by create at least one virtual machine in this locality, realizes general-purpose computations task by operation virtual machine.
On the basis of embodiment tetra-, in another embodiment, after step 402, it is the calculation task that client cloud management platform sends, by operation virtual machine, utilize after computational resource completes calculation task, also comprise: the resource release request that client cloud management platform sends, according to resource release request computational resource.After calculation task completes, by timely computational resource being discharged, on the one hand can reduce user's cost, on the other hand, by by the resource of release for other users, also can improve the utilization factor of resource.
Because it is insecure that the individual PC resource of computational resource is provided, it is much higher that the likelihood ratio that break down or network disconnection etc. causes service to use plays self-built data center of operator.How in insecure resource, to realize reliably and calculating, be a key issue.In order to improve the reliability of result of calculation, if just calculation task needs M step can complete calculating, a checkpoint is set after each step; When client moves to checkpoint, the result of calculation of this step is reported to cloud management platform, so that management platform is according to the result of calculation of the result of calculation of this client and other clients, by arbitration mechanism, guarantee the correctness of result of calculation, specific implementation just repeats no more here, can be with reference to the description of embodiment bis-and embodiment tri-.
In actual use procedure, the computational resource that user need use according to reality is paid, and toll-ratio is formulated by operator, supposes to using that a computing unit of every use (CU) is designated as basic charge unit (1cuhour) computing time of one hour.The cuhour number that user need use according to reality is paid, and toll-ratio is formulated by operator, is designated as f1.Meanwhile, operator is according to the actual use of the cuhour number of supplier's account number record, and Monthly Payment is to supplier, and toll-ratio is formulated by operator, is designated as f2.Operator is f1-f2 by the gross profit of hiring out each cuhour and obtaining.Supplier, for many profits, need to possess high bandwidth between possible user and other suppliers, and cuhour as much as possible is provided.In a lot of situations, supplier is personal user, just utilizes idle personal computer and broadband to obtain some incomes, almost there is no expense.The operation cost of operator is mainly the normal operation that maintains cloud management platform, does not need to drop into huge capital and sets up and safeguard large-scale data center.And user can obtain elasticity computing power with the cost more much lower than other cloud computing provider on market (IaaS cloud provider).
Fig. 6 is the structural representation of cloud management platform embodiment mono-of the present invention, and as shown in Figure 6, the cloud management platform of the present embodiment can comprise: receiver module 51, resource determination module 52, sending module 53, result of calculation determination module 54.
Wherein, receiver module 51, the cloud resource request sending for receiving subscriber equipment, comprises calculation task and completes the required computational resource information of calculation task in cloud resource request;
Sending module 53, for calculation task is sent at least one of computational resource client is provided, makes each client by each virtual machine of operation, complete calculation task respectively, and returns to cloud management platform the result of calculation that calculation task is corresponding;
Result of calculation determination module 54, determines for the result of calculation receiving according to receiver module the final calculation result that calculation task is corresponding.
In the present embodiment, resource determination module 52 specifically for: according to computational resource information, and subscriber equipment and the Internet resources between each client of idle computational resource are provided, from idling-resource, determine the computational resource into user equipment allocation.
The cloud management platform that the present embodiment provides also comprises: resource release module, correspondingly, receiver module 51 also for: receive the resource release request that subscriber equipment sends; Resource release module, discharges for the resource release request receiving according to receiver module 51 computational resource that at least one client provides.
In the present embodiment, receiver module 51 also for: receive the idle computational resource information that each client reports.Idle computational resource information at least comprises: the surplus capacity of the CPU of each client, free memory space, remanence disk space.By the idle computational resource of each client of cloud management platform unified management, and it is user assignment computational resource.
The cloud management platform that the present embodiment provides, can be used for the technical scheme that manner of execution embodiment mono-provides, and specific implementation and technique effect are similar, repeat no more here.
Fig. 7 is the structural representation of cloud management platform embodiment bis-of the present invention, and as shown in Figure 7, the cloud management platform of the present embodiment comprises: receiver module 61, resource determination module 62, sending module 63, result of calculation determination module 64.
Wherein, receiver module 61, the cloud resource request sending for receiving subscriber equipment, comprises calculation task and completes the required computational resource information of calculation task in cloud resource request;
Sending module 63, for calculation task is sent at least one of computational resource client is provided, makes each client by each virtual machine of operation, complete calculation task respectively, and returns to cloud management platform the result of calculation that calculation task is corresponding;
Result of calculation determination module 64, determines for the result of calculation receiving according to receiver module the final calculation result that calculation task is corresponding.
In the present embodiment, the computing unit tolerance of computational resource after with standardization, each computing unit has identical CPU processing power and memory headroom, disk space, in cloud resource request, comprise to complete the required computational resource information of calculation task be the number N of the required computing unit of calculation task; Resource determination module 62 specifically for: according to the number N that completes the required computing unit of calculation task, from idle computing unit, determine N the computing unit into user equipment allocation.
In order to guarantee the reliability of result of calculation, in the present embodiment, in cloud resource request, also comprise: redundant resource sign, redundant resource sign is used for to cloud management platform application three batch total operator resources, every batch total operator resource comprises respectively N computing unit, and every batch total operator resource has been respectively used to calculation task; Accordingly, resource determination module 62 specifically for: according to redundant computation sign and computational resource information, from idle computing unit, determine the three batch total operator resources into user equipment allocation;
Accordingly, sending module 63 specifically for: calculation task is sent to respectively to each client that three batch total operator resources are provided, so that each virtual machine moving in each client calculates respectively calculation task, and return to cloud management platform three result of calculations that calculation task is corresponding; Receiver module 61 specifically for: receive three result of calculations corresponding to calculation task; Result of calculation determination module 64 specifically for: according to three result of calculations that calculation task is corresponding, determine the final calculation result that calculation task is corresponding.
If calculation task needs M step just can complete calculating, a checkpoint is set after each step, checkpoint is for when each client moves to checkpoint, and each client reports to cloud management platform by the result of calculation of this step separately respectively; Correspondingly, result of calculation determination module specifically for: the result of calculation of this step separately reporting respectively according to the client formerly receiving in two batch total operator resources, judges that whether the result of calculation of this step that two batch total operator resources report identical;
If, the result of calculation that formerly receives this step separately that the client in two batch total operator resources reports is respectively identical, the identical result of calculation two batch total operator resources being reported is as the final calculation result of this step, if management platform is not also received the result of calculation of this step that another batch total operator resource reports, stop the operation of another batch total operator resource.
If not, the result of calculation that formerly receives this step separately that the client in two batch total operator resources reports is respectively identical, wait for the result of calculation of this step that another batch total operator resource reports, whether judge in the result of calculation of this step that three batch total operator resources report has two batch totals to come to the same thing, compare between two, the result of calculation that judges whether two groups is identical, if have two batch totals to come to the same thing in the result of calculation of this step that three batch total operator resources report, the identical result of calculation two batch total operator resources being reported is as the final calculation result of this step.
Then, at synchronization, using the final calculation result of this step as next step initial value, send to respectively each client of three batch total operator resources, so that each client of three batch total operator resources starts next step according to the final calculation result of this step simultaneously, calculate, until M step finishes; Using the net result of M step as final calculation result corresponding to calculation task.
In the present embodiment, if three result of calculations that in certain step, three batch total operator resources report are all different, result of calculation determination module
64specifically for: the calculation task that finishes three batch total operator resources.
The cloud management platform of the present embodiment, the technical scheme that can provide for manner of execution embodiment bis-and embodiment tri-, it realizes principle and technique effect is similar, repeats no more herein.
Fig. 8 is the structural representation of client implementation example one of the present invention, and as shown in Figure 8, the client of the present embodiment comprises: virtual machine creating module 71, receiver module 72, computing module 73 and sending module 74.
Wherein, virtual machine creating module 71, for create at least one virtual machine in this locality, each virtual machine is for providing computational resource to cloud management platform;
Sending module 74, for result of calculation corresponding to calculation task is sent to cloud management platform, so that cloud management platform is determined the final calculation result of calculation task according to result of calculation corresponding to calculation task.
In the present embodiment, virtual machine creating module 71 specifically for: from cloud management platform, download cloud service plug-in unit, at local runtime cloud service plug-in unit, by cloud service plug-in unit, in this locality, create at least one virtual machine.
The client of the present embodiment, can also comprise: idling-resource reporting module, for obtaining utilization factor, disk utilization, the memory usage of local cpu, determine local idle computational resource, to cloud management platform, report idle computational resource, idle computational resource information at least comprises: the surplus capacity of the CPU of each client, free memory space, remanence disk space.
Further, client can also comprise: resource release module, receiver module 72 also for: receive the resource release request that cloud management platform sends; Resource release module, for discharging computational resource according to resource release request.
In a kind of optional implementation, computational resource can be with the computing unit tolerance after standardization, and each computing unit position has identical CPU processing power, memory headroom and disk space; Correspondingly, idling-resource reporting module is specifically for number from idle computing unit to cloud management platform that report.
In order to guarantee the accuracy of result of calculation, if just calculation task needs M step can complete calculating, a checkpoint to be set after each step; Sending module 74 also for: when moving to checkpoint, the result of calculation of this step is reported to cloud management platform, so that cloud management platform is according to the result of calculation of this step, determine the final calculation result of this step, using the final calculation result of this step as next step initial value, send to client, accordingly, next step the initial value that the receiver module 71 of client receives that cloud management platforms send.
The client that the present embodiment provides, can be used for the technical scheme that manner of execution embodiment tetra-provides, and specific implementation and technique effect type, repeat no more here.
Fig. 9 is the structural representation of cloud management platform embodiment tri-of the present invention, and as shown in Figure 9, the cloud management platform 800 that the present embodiment provides comprises: processor 81, storer 82, transmitter 83, receiver 84.Storer 82, transmitter 83 and receiver 84 are connected with processor 81 by bus.Wherein, instructions are carried out in storer 82 storage, when cloud management platform 800 received shipments are when capable, between processor 81 and storer 82, communicate by letters, and processor 81 execution instructions make cloud management platform 800 execution cloud computing method provided by the invention.
Wherein, the cloud resource request that receiver 84. sends for receiving subscriber equipment, comprises calculation task and completes the required computational resource information of calculation task in cloud resource request;
In the present embodiment, processor 81 can be determined the computational resource of distributing to client in the following manner from idling-resource, particularly, processor 81 is according to computational resource information, and subscriber equipment and the Internet resources between each client of idle computational resource are provided, from idling-resource, determine the computational resource into user equipment allocation.Idling-resource is that the idle computational resource information that reports according to client is determined, therefore, receiver 84 also for: receive the idle computational resource information that each client reports.Wherein, idle computational resource information at least comprises: the surplus capacity of the CPU of each client, free memory space, remanence disk space.
After completing calculation task, receiver 84 also for: receive the resource release request that subscriber equipment sends; Then, processor 81 discharges according to resource release request the computational resource that at least one client provides.
In a kind of implementation of the present embodiment, computational resource can be with the computing unit tolerance after standardization, each computing unit has identical CPU processing power and memory headroom, disk space, in cloud resource request, comprise to complete the required computational resource information of calculation task be the number N of the required computing unit of calculation task.Processor 81 specifically for: according to the number N that completes the required computing unit of calculation task, from idle computing unit, determine N the computing unit into user equipment allocation.
In order to guarantee the accuracy of result of calculation, in cloud resource request, also comprise: redundant resource sign, redundant resource sign is used for to cloud management platform application three batch total operator resources, and every batch total operator resource comprises respectively N computing unit, and every batch total operator resource has been respectively used to calculation task; By application redundant computation resource, jointly complete calculation task, thereby guarantee the reliable of checkout result.Processor 81, specifically for according to redundant computation sign and computational resource information, is determined the three batch total operator resources into user equipment allocation from idle computing unit.
Then, transmitter 83 sends to calculation task respectively each client that three batch total operator resources are provided, so that each virtual machine moving in each client calculates respectively calculation task, and return to cloud management platform three result of calculations that calculation task is corresponding, accordingly, receiver 84 receives three result of calculations corresponding to calculation task that client sends; Processor 81 determines according to three result of calculations corresponding to calculation task the final calculation result that calculation task is corresponding.
In this enforcement, if calculation task needs M step just can complete calculating, a checkpoint is set after each step, checkpoint is for when each client moves to checkpoint, and each client reports to cloud management platform by the result of calculation of this step separately respectively;
Correspondingly, processor 81, according to the result of calculation that formerly receives this step separately that the client in two batch total operator resources reports respectively, judges that whether the result of calculation of this step that two batch total operator resources report is identical.If so, the identical result of calculation two batch total operator resources being reported, as the final calculation result of this step, if management platform is not also received the result of calculation of this step that another batch total operator resource reports, stops the operation of another batch total operator resource; If not, wait for the result of calculation of this step that another batch total operator resource reports, judge in the result of calculation of this step that three batch total operator resources report and whether have two batch totals to come to the same thing.
If have two batch totals to come to the same thing in the result of calculation of this step that three batch total operator resources report, the identical result of calculation two batch total operator resources being reported is as the final calculation result of this step; Then, at synchronization, using the final calculation result of this step as next step initial value, by transmitter 83, send to respectively each client of three batch total operator resources, so that starting next step according to the final calculation result of this step simultaneously, each client of three batch total operator resources calculates, until M step finishes; Using the net result of M step as final calculation result corresponding to calculation task.
If three result of calculations that in certain step, three batch total operator resources report are all different, result of calculation determination module is specifically for the calculation task that finishes three batch total operator resources.
The cloud management platform that the present embodiment provides, can be used for the technical scheme of manner of execution embodiment mono-to three, and specific implementation and technique effect are similar, repeat no more here.
Figure 10 is the structural representation of client implementation example two of the present invention, and as shown in figure 10, the client 900 that the present embodiment provides comprises: processor 91, storer 92, transmitter 93, receiver 94.Storer 92, transmitter 93 and receiver 94 are connected with processor 91 by bus.Wherein, instructions are carried out in storer 92 storage, when client 900 operation, between processor 91 and storer 92, communicate by letters, and processor 91 execution instructions make client 900 execution cloud computing method provided by the invention.
Wherein, processor 91, for create at least virtual machine in this locality, each virtual machine is for providing computational resource to cloud management platform;
In the present embodiment, receiver 94 also for: receive the resource release request that cloud management platform sends; Processor 91 is for discharging computational resource according to resource release request.
In a kind of feasible implementation, the computing unit tolerance of computational resource after with standardization, each computing unit position has identical CPU processing power, memory headroom and disk space; Correspondingly, transmitter 93 is specifically for reporting the number of idle computing unit to cloud management platform.
If calculation task needs M step just can complete calculating, a checkpoint is set after each step; Transmitter 93 specifically for: when moving to checkpoint, the result of calculation of this step is reported to cloud management platform, so that cloud management platform is according to the result of calculation of this step, determine the final calculation result of this step, using the final calculation result of this step as next step initial value, send to client; Accordingly, corresponding, receiver 94 is also for, the initial value of next step that receives that cloud management platform sends.
The client of the present embodiment, the technical scheme that can provide for manner of execution embodiment tetra-, it realizes principle and technique effect is similar, repeats no more herein.
One of ordinary skill in the art will appreciate that: all or part of step that realizes above-mentioned each embodiment of the method can complete by the relevant hardware of programmed instruction.Aforesaid program can be stored in a computer read/write memory medium.This program, when carrying out, is carried out the step that comprises above-mentioned each embodiment of the method; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CDs.
Finally it should be noted that: each embodiment, only in order to technical scheme of the present invention to be described, is not intended to limit above; Although the present invention is had been described in detail with reference to aforementioned each embodiment, those of ordinary skill in the art is to be understood that: its technical scheme that still can record aforementioned each embodiment is modified, or some or all of technical characterictic is wherein equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.
Claims (30)
1. a cloud computing method, is characterized in that, comprising:
Cloud management platform receives the cloud resource request that subscriber equipment sends, and comprises calculation task and complete the required computational resource information of described calculation task in described cloud resource request;
Described cloud management platform is according to the described computational resource information comprising in described cloud resource request, from idle computational resource, determine the computational resource into described user equipment allocation, described computational resource is provided by least one client, and in each client, operation has at least one virtual machine;
Described cloud management platform described calculation task is provided provide at least one of described computational resource client, make described each client by described each virtual machine of operation, complete described calculation task respectively, and return to described cloud management platform the result of calculation that described calculation task is corresponding;
Described cloud management platform receives the result of calculation that described calculation task is corresponding, according to described result of calculation, determines the final calculation result that described calculation task is corresponding.
2. method according to claim 1, is characterized in that, described cloud management platform, according to described computational resource information, is determined the computational resource into described user equipment allocation from idle computational resource, comprising:
Described cloud management platform is according to described computational resource information, and described subscriber equipment and the Internet resources between each client of described idle computational resource are provided, and determines the described computational resource into described user equipment allocation from described idling-resource.
3. method according to claim 1, is characterized in that, described cloud management platform receives described result of calculation, after determining final calculation result, also comprises according to described result of calculation:
Described cloud management platform receives the resource release request that described subscriber equipment sends;
Described cloud management platform discharges according to described resource release request the described computational resource that described at least one client provides.
4. method according to claim 1, is characterized in that, also comprises:
Described cloud management platform receives the idle computational resource information that described each client reports.
5. method according to claim 4, is characterized in that, described idle computational resource information at least comprises: the surplus capacity of the central processing unit CPU of described each client, free memory space, remanence disk space.
6. according to the method described in claim 1-5 any one, it is characterized in that, the computing unit tolerance of described computational resource after with standardization, each computing unit has identical CPU processing power and memory headroom, disk space, in described cloud resource request, comprise to complete the required computational resource information of described calculation task be the number N of the required computing unit of described calculation task;
Described cloud management platform, according to described computational resource information, is determined the computational resource into described user equipment allocation from idle computational resource, comprising:
Described cloud management platform, according to the number N that completes the required computing unit of described calculation task, is determined N the computing unit into described user equipment allocation from idle computing unit.
7. method according to claim 6, is characterized in that, in described cloud resource request, also comprises:
Redundant resource sign, described redundant resource sign is used for to described cloud management platform application three batch total operator resources, and described every batch total operator resource comprises respectively N computing unit, and described every batch total operator resource has been respectively used to described calculation task;
Accordingly, described cloud management platform, according to described computational resource information, is determined as described user equipment allocation computational resource from idle computational resource, comprising:
Described cloud management platform, according to described redundant computation sign and described computational resource information, is determined the three batch total operator resources into described user equipment allocation from described idle computing unit;
Described cloud management platform described calculation task is provided provide at least one of described computational resource client, make described each client by described each virtual machine of operation, complete described calculation task respectively, and return to described cloud management platform the result of calculation that described calculation task is corresponding, comprising:
Described cloud management platform sends to described calculation task respectively each client that described three batch total operator resources are provided, so that each virtual machine moving in described each client calculates respectively described calculation task, and return to described cloud management platform three result of calculations that described calculation task is corresponding;
Described cloud management platform receives the result of calculation that described calculation task is corresponding, according to described result of calculation, determines the final calculation result that described calculation task is corresponding, comprising:
Described cloud management platform receives three result of calculations that described calculation task is corresponding, according to three result of calculations that described calculation task is corresponding, determines the final calculation result that described calculation task is corresponding.
8. method according to claim 7, it is characterized in that, if described calculation task needs M step just can complete calculating, a checkpoint is set after each step, described checkpoint is for when described each client moves to described checkpoint, and described each client reports to described cloud management platform by the result of calculation of this step separately respectively;
Correspondingly, described cloud management platform receives three result of calculations that described calculation task is corresponding, according to three result of calculations that described calculation task is corresponding, determines the final calculation result that described calculation task is corresponding, comprising:
Described cloud management platform, according to the result of calculation that formerly receives this step separately that the client in two batch total operator resources reports respectively, judges that whether the result of calculation of this step that described two batch total operator resources report is identical;
If, the identical result of calculation that described cloud management platform reports described two batch total operator resources is as the final calculation result of this step, if described management platform is not also received the result of calculation of this step that another batch total operator resource reports, stop the operation of described another batch total operator resource;
Whether if not, described cloud management platform is waited for the result of calculation of this step that described another batch total operator resource reports, judge in the result of calculation of this step that described three batch total operator resources report and have two batch totals to come to the same thing;
If have two batch totals to come to the same thing in the result of calculation of this step that described three batch total operator resources report, the identical result of calculation that described cloud management platform reports described two batch total operator resources is as the final calculation result of this step;
Described cloud management platform is at synchronization, using the final calculation result of described step as next step initial value, send to respectively each client of described three batch total operator resources, so that starting next step according to the final calculation result of described step simultaneously, each client of described three batch total operator resources calculates, until M step finishes;
Described cloud management platform is using the net result of described M step as final calculation result corresponding to described calculation task.
9. method according to claim 8, is characterized in that, if three result of calculations that in certain step, three batch total operator resources report are all different, described management platform finishes the calculation task of described three batch total operator resources.
10. a cloud computing method, is characterized in that, comprising:
Client creates at least one virtual machine in this locality, described each virtual machine is for providing computational resource to cloud management platform;
The calculation task that described in described client, cloud management platform sends, by moving described each virtual machine, utilize described computational resource to complete described calculation task, and result of calculation corresponding to described calculation task is sent to described cloud management platform, so that described cloud management platform is determined the final calculation result of described calculation task according to result of calculation corresponding to described calculation task.
11. methods according to claim 10, is characterized in that, described client creates at least one virtual machine in this locality, comprising:
Described client is downloaded cloud service plug-in unit from described cloud management platform, at cloud service plug-in unit described in local runtime, creates described at least one virtual machine by described cloud service plug-in unit in this locality.
12. methods according to claim 10, is characterized in that, also comprise:
Described client is obtained utilization factor, disk utilization, the memory usage of local central processing unit CPU, determine local idle computational resource, to described cloud management platform, report described idle computational resource, idle computational resource information at least comprises: the surplus capacity of the CPU of described each client, free memory space, remanence disk space.
13. methods according to claim 10, is characterized in that, the calculation task that cloud management platform sends described in described client utilizes after described computational resource completes described calculation task by moving described virtual machine, also comprises:
The resource release request that described in described client, cloud management platform sends, discharges described computational resource according to described resource release request.
14. according to the method described in claim 10-13 any one, it is characterized in that, and the computing unit tolerance of described computational resource after with standardization, each computing unit position has identical CPU processing power, memory headroom and disk space;
Correspondingly, described client reports described idle computational resource to described cloud management platform, comprising:
Described client reports the number of idle computing unit to described cloud management platform.
15. methods according to claim 14, is characterized in that, if described calculation task needs M step just can complete calculating, a checkpoint are set after each step;
When described client moves to described checkpoint, the result of calculation of this step is reported to described cloud management platform, so that described cloud management platform is according to the result of calculation of described step, determine the final calculation result of this step, using the final calculation result of this step as next step initial value, send to described client;
Accordingly, next step the initial value that described client also receives that described cloud management platform sends.
16. 1 kinds of cloud management platforms, is characterized in that, comprising:
Receiver module, the cloud resource request sending for receiving subscriber equipment, comprises calculation task in described cloud resource request and completes the required computational resource information of described calculation task;
Resource determination module, for the described computational resource information comprising according to described cloud resource request, from idle computational resource, determine the computational resource into described user equipment allocation, described computational resource is provided by least one client, and in each client, operation has at least one virtual machine;
Sending module, for described calculation task is sent at least one of described computational resource client is provided, make described each client by described each virtual machine of operation, complete described calculation task respectively, and return to described cloud management platform the result of calculation that described calculation task is corresponding;
Described receiver module, also for receiving the result of calculation that described calculation task is corresponding;
Result of calculation determination module, determines for the described result of calculation receiving according to described receiver module the final calculation result that described calculation task is corresponding.
17. cloud management platforms according to claim 16, is characterized in that, described resource determination module specifically for:
According to described computational resource information, and described subscriber equipment and the Internet resources between each client of described idle computational resource are provided, from described idling-resource, determine the described computational resource into described user equipment allocation.
18. cloud management platforms according to claim 16, is characterized in that, also comprise: resource release module;
Described receiver module also for: receive the resource release request that described subscriber equipment sends;
Described resource release module, for discharging the described computational resource that described at least one client provides according to described resource release request.
19. cloud management platforms according to claim 16, is characterized in that, described receiver module also for:
Receive the idle computational resource information that described each client reports.
20. cloud management platforms according to claim 19, is characterized in that, described idle computational resource information at least comprises: the surplus capacity of the central processing unit CPU of described each client, free memory space, remanence disk space.
21. according to cloud management platform described in claim 16-20 any one, it is characterized in that, the computing unit tolerance of described computational resource after with standardization, each computing unit has identical CPU processing power and memory headroom, disk space, in described cloud resource request, comprise to complete the required computational resource information of described calculation task be the number N of the required computing unit of described calculation task;
Described resource determination module specifically for: according to the number N that completes the required computing unit of described calculation task, from idle computing unit, determine N the computing unit into described user equipment allocation.
22. cloud management platforms according to claim 21, is characterized in that, in described cloud resource request, also comprise:
Redundant resource sign, described redundant resource sign is used for to described cloud management platform application three batch total operator resources, and described every batch total operator resource comprises respectively N computing unit, and described every batch total operator resource has been respectively used to described calculation task;
Accordingly, described resource determination module specifically for: according to described redundant computation sign and described computational resource information, from described idle computing unit, determine the three batch total operator resources into described user equipment allocation;
Described sending module specifically for: described calculation task is sent to respectively to each client that described three batch total operator resources are provided, so that each virtual machine moving in described each client calculates respectively described calculation task, and return to described cloud management platform three result of calculations that described calculation task is corresponding;
Described receiver module specifically for: receive three result of calculations that described calculation task is corresponding;
Described result of calculation determination module specifically for: according to three result of calculations that described calculation task is corresponding, determine the final calculation result that described calculation task is corresponding.
23. cloud management platforms according to claim 22, it is characterized in that, if described calculation task needs M step just can complete calculating, a checkpoint is set after each step, described checkpoint is for when described each client moves to described checkpoint, and described each client reports to described cloud management platform by the result of calculation of this step separately respectively;
Correspondingly, result of calculation determination module specifically for: the result of calculation of this step separately reporting respectively according to the client formerly receiving in two batch total operator resources, judges that whether the result of calculation of this step that described two batch total operator resources report identical;
If, the identical result of calculation described two batch total operator resources being reported is as the final calculation result of this step, if described management platform is not also received the result of calculation of this step that another batch total operator resource reports, stop the operation of described another batch total operator resource;
If not, wait for the result of calculation of this step that described another batch total operator resource reports, judge in the result of calculation of this step that described three batch total operator resources report and whether have two batch totals to come to the same thing;
If have two batch totals to come to the same thing in the result of calculation of this step that described three batch total operator resources report, the identical result of calculation described two batch total operator resources being reported is as the final calculation result of this step;
At synchronization, using the final calculation result of described step as next step initial value, send to respectively each client of described three batch total operator resources, so that starting next step according to the final calculation result of described step simultaneously, each client of described three batch total operator resources calculates, until M step finishes;
Using the net result of described M step as final calculation result corresponding to described calculation task.
24. cloud management platforms according to claim 23, it is characterized in that, if three result of calculations that in certain step, three batch total operator resources report are all different, described result of calculation determination module is specifically for the calculation task that finishes described three batch total operator resources.
25. 1 kinds of clients, is characterized in that, comprising:
Virtual machine creating module, for create at least virtual machine in this locality, described each virtual machine is for providing computational resource to cloud management platform;
Receiver module, the calculation task sending for receiving described cloud management platform;
Computing module, for utilizing described computational resource to complete described calculation task by moving described each virtual machine;
Sending module, for result of calculation corresponding to described calculation task is sent to described cloud management platform, so that described cloud management platform is determined the final calculation result of described calculation task according to result of calculation corresponding to described calculation task.
26. clients according to claim 25, is characterized in that, described virtual machine creating module specifically for:
From described cloud management platform, download cloud service plug-in unit, at cloud service plug-in unit described in local runtime, by described cloud service plug-in unit, in this locality, create described at least one virtual machine.
27. clients according to claim 25, is characterized in that, also comprise:
Idling-resource reporting module, for obtaining utilization factor, disk utilization, the memory usage of local central processing unit CPU, determine local idle computational resource, to described cloud management platform, report described idle computational resource, idle computational resource information at least comprises: the surplus capacity of the CPU of described each client, free memory space, remanence disk space.
28. clients according to claim 25, is characterized in that, also comprise: resource release module;
Described receiver module also for: receive the resource release request that described cloud management platform sends;
Described resource release module, for discharging described computational resource according to described resource release request.
29. according to the client described in claim 25-28 any one, it is characterized in that, and the computing unit tolerance of described computational resource after with standardization, each computing unit position has identical CPU processing power, memory headroom and disk space;
Correspondingly, described idling-resource reporting module is specifically for number from idle computing unit to described cloud management platform that report.
30. clients according to claim 29, is characterized in that, if described calculation task needs M step just can complete calculating, a checkpoint are set after each step;
Described sending module specifically for: when moving to described checkpoint, the result of calculation of this step is reported to described cloud management platform, so that described cloud management platform is according to the result of calculation of described step, determine the final calculation result of this step, using the final calculation result of this step as next step initial value, send to described client;
Accordingly, described receiver module is also for, the initial value of next step that receives that described cloud management platform sends.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310517736.1A CN103544064B (en) | 2013-10-28 | 2013-10-28 | Cloud computing method, cloud management platform and client |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310517736.1A CN103544064B (en) | 2013-10-28 | 2013-10-28 | Cloud computing method, cloud management platform and client |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103544064A true CN103544064A (en) | 2014-01-29 |
CN103544064B CN103544064B (en) | 2018-03-13 |
Family
ID=49967539
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310517736.1A Expired - Fee Related CN103544064B (en) | 2013-10-28 | 2013-10-28 | Cloud computing method, cloud management platform and client |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103544064B (en) |
Cited By (34)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103823719A (en) * | 2014-02-26 | 2014-05-28 | 杭州群核信息技术有限公司 | Distributed cloud computing system and distributed cloud computing method for executable program |
CN104125297A (en) * | 2014-08-06 | 2014-10-29 | 华为技术有限公司 | Virtual resource sharing method, device and system |
CN104202436A (en) * | 2014-09-28 | 2014-12-10 | 北京奇虎科技有限公司 | Method and device for disposing service request through advertisement platform |
CN104506600A (en) * | 2014-12-16 | 2015-04-08 | 苏州海博智能系统有限公司 | Computation resource sharing method, device and system as well as client side and server |
CN105187483A (en) * | 2015-07-29 | 2015-12-23 | 北京中油瑞飞信息技术有限责任公司 | Method and device for allocating cloud computing resources |
CN106255957A (en) * | 2014-04-30 | 2016-12-21 | 瑞典爱立信有限公司 | The distribution of cloud computing resources |
CN106412094A (en) * | 2016-11-02 | 2017-02-15 | 深圳前海生生科技有限公司 | A method for organizing and managing scattered resources in a public cloud mode |
CN106502796A (en) * | 2016-10-27 | 2017-03-15 | 广东浪潮大数据研究有限公司 | A kind of software development test platform, system and using method |
CN106712998A (en) * | 2015-11-18 | 2017-05-24 | 中兴通讯股份有限公司 | Cloud platform resource management method, device and system |
CN107006068A (en) * | 2015-03-27 | 2017-08-01 | 华为技术有限公司 | A kind of wireless access network equipment, collocation method and configuration device |
CN107040406A (en) * | 2017-03-14 | 2017-08-11 | 西安电子科技大学 | A kind of end cloud cooperated computing system and its fault-tolerance approach |
CN107209697A (en) * | 2015-02-03 | 2017-09-26 | 戴尔产品有限公司 | The workload of dynamic control is performed |
CN107343017A (en) * | 2016-12-21 | 2017-11-10 | 北京安天网络安全技术有限公司 | For managing the method, service end and terminal of GPU resource |
CN107608795A (en) * | 2017-09-19 | 2018-01-19 | 百度在线网络技术(北京)有限公司 | cloud computing method and device |
CN108009010A (en) * | 2017-12-01 | 2018-05-08 | 携程旅游信息技术(上海)有限公司 | The managing device of thin-client, system, method, electronic equipment, storage medium |
CN108958935A (en) * | 2018-06-28 | 2018-12-07 | 阿瑞思科技(成都)有限责任公司 | The method and its system of distributed computing are realized based on mobile terminal |
CN109040206A (en) * | 2018-07-09 | 2018-12-18 | 广东工业大学 | A kind of computing resource long-term allocation method based on mobile device cloud |
CN109246232A (en) * | 2018-09-30 | 2019-01-18 | 西安万像电子科技有限公司 | Data transmission method, system and device |
CN109729106A (en) * | 2017-10-27 | 2019-05-07 | 伊姆西Ip控股有限责任公司 | Handle the method, system and computer program product of calculating task |
CN109766193A (en) * | 2019-01-29 | 2019-05-17 | 努比亚技术有限公司 | Calculating task executes method, server, terminal and computer readable storage medium |
CN110027596A (en) * | 2019-03-29 | 2019-07-19 | 北京交通大学 | A kind of Introduction of Train Operation Control System based on cloud computing |
CN110213180A (en) * | 2019-06-06 | 2019-09-06 | 腾讯科技(深圳)有限公司 | Method for managing network resource, device and cloud platform |
WO2019238092A1 (en) * | 2018-06-15 | 2019-12-19 | 华为技术有限公司 | Management method, device and system for virtual machine |
CN110851529A (en) * | 2019-11-01 | 2020-02-28 | 腾讯科技(深圳)有限公司 | Computing power scheduling method and related equipment |
CN111461884A (en) * | 2020-03-31 | 2020-07-28 | 杭州溪塔科技有限公司 | Trusted computing service sharing method, device and system based on block chain |
CN111565101A (en) * | 2020-04-28 | 2020-08-21 | 上海物融智能科技有限公司 | Processing method and device for computing task |
CN112068954A (en) * | 2020-08-18 | 2020-12-11 | 弥伦工业产品设计(上海)有限公司 | Method and system for scheduling network computing resources |
CN112106418A (en) * | 2018-03-20 | 2020-12-18 | 诺基亚技术有限公司 | Apparatus, method and computer program |
CN112148439A (en) * | 2019-06-28 | 2020-12-29 | 浙江宇视科技有限公司 | Task processing method, device, equipment and storage medium |
CN112274914A (en) * | 2020-10-29 | 2021-01-29 | 西安万像电子科技有限公司 | Method for realizing exclusive game, device for realizing exclusive game and cloud game system |
CN112400301A (en) * | 2018-07-31 | 2021-02-23 | 韦斯特尔电子工业和贸易有限责任公司 | Method, apparatus, system and computer program for data distribution |
CN112416592A (en) * | 2020-11-26 | 2021-02-26 | 张新利 | Method for providing computing power of cloud server |
CN114095577A (en) * | 2020-07-31 | 2022-02-25 | 中国移动通信有限公司研究院 | Resource request method and device, calculation network element node and calculation application equipment |
CN115378829A (en) * | 2021-05-17 | 2022-11-22 | 中国移动通信有限公司研究院 | Resource request method and device, computational power network element node and terminal equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102769615A (en) * | 2012-07-02 | 2012-11-07 | 北京大学 | Task scheduling method and system based on MapReduce mechanism |
CN103064728A (en) * | 2012-12-13 | 2013-04-24 | 河南工业大学 | Fault-tolerant scheduling method of Map Reduce task |
CN103309745A (en) * | 2013-04-19 | 2013-09-18 | 无锡成电科大科技发展有限公司 | Method and device for distributing virtual resource in cloud architecture |
CN103365713A (en) * | 2012-04-01 | 2013-10-23 | 华为技术有限公司 | Resource dispatch and management method and device |
-
2013
- 2013-10-28 CN CN201310517736.1A patent/CN103544064B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103365713A (en) * | 2012-04-01 | 2013-10-23 | 华为技术有限公司 | Resource dispatch and management method and device |
CN102769615A (en) * | 2012-07-02 | 2012-11-07 | 北京大学 | Task scheduling method and system based on MapReduce mechanism |
CN103064728A (en) * | 2012-12-13 | 2013-04-24 | 河南工业大学 | Fault-tolerant scheduling method of Map Reduce task |
CN103309745A (en) * | 2013-04-19 | 2013-09-18 | 无锡成电科大科技发展有限公司 | Method and device for distributing virtual resource in cloud architecture |
Cited By (46)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103823719A (en) * | 2014-02-26 | 2014-05-28 | 杭州群核信息技术有限公司 | Distributed cloud computing system and distributed cloud computing method for executable program |
CN106255957A (en) * | 2014-04-30 | 2016-12-21 | 瑞典爱立信有限公司 | The distribution of cloud computing resources |
CN104125297A (en) * | 2014-08-06 | 2014-10-29 | 华为技术有限公司 | Virtual resource sharing method, device and system |
CN104202436A (en) * | 2014-09-28 | 2014-12-10 | 北京奇虎科技有限公司 | Method and device for disposing service request through advertisement platform |
CN104506600A (en) * | 2014-12-16 | 2015-04-08 | 苏州海博智能系统有限公司 | Computation resource sharing method, device and system as well as client side and server |
CN107209697B (en) * | 2015-02-03 | 2022-06-07 | 戴尔产品有限公司 | Dynamically controlled workload execution |
CN107209697A (en) * | 2015-02-03 | 2017-09-26 | 戴尔产品有限公司 | The workload of dynamic control is performed |
CN107006068A (en) * | 2015-03-27 | 2017-08-01 | 华为技术有限公司 | A kind of wireless access network equipment, collocation method and configuration device |
CN107006068B (en) * | 2015-03-27 | 2020-09-11 | 华为技术有限公司 | Wireless access network equipment, configuration method and configuration device |
CN105187483A (en) * | 2015-07-29 | 2015-12-23 | 北京中油瑞飞信息技术有限责任公司 | Method and device for allocating cloud computing resources |
CN105187483B (en) * | 2015-07-29 | 2018-05-08 | 北京中油瑞飞信息技术有限责任公司 | Distribute the method and device of cloud computing resources |
WO2017084453A1 (en) * | 2015-11-18 | 2017-05-26 | 中兴通讯股份有限公司 | Cloud platform resource management method, device and system |
CN106712998A (en) * | 2015-11-18 | 2017-05-24 | 中兴通讯股份有限公司 | Cloud platform resource management method, device and system |
CN106502796A (en) * | 2016-10-27 | 2017-03-15 | 广东浪潮大数据研究有限公司 | A kind of software development test platform, system and using method |
CN106412094A (en) * | 2016-11-02 | 2017-02-15 | 深圳前海生生科技有限公司 | A method for organizing and managing scattered resources in a public cloud mode |
CN107343017A (en) * | 2016-12-21 | 2017-11-10 | 北京安天网络安全技术有限公司 | For managing the method, service end and terminal of GPU resource |
CN107040406A (en) * | 2017-03-14 | 2017-08-11 | 西安电子科技大学 | A kind of end cloud cooperated computing system and its fault-tolerance approach |
CN107040406B (en) * | 2017-03-14 | 2020-08-11 | 西安电子科技大学 | End cloud cooperative computing system and fault-tolerant method thereof |
CN107608795A (en) * | 2017-09-19 | 2018-01-19 | 百度在线网络技术(北京)有限公司 | cloud computing method and device |
CN109729106A (en) * | 2017-10-27 | 2019-05-07 | 伊姆西Ip控股有限责任公司 | Handle the method, system and computer program product of calculating task |
CN109729106B (en) * | 2017-10-27 | 2021-07-16 | 伊姆西Ip控股有限责任公司 | Method, system and computer program product for processing computing tasks |
CN108009010B (en) * | 2017-12-01 | 2021-03-23 | 携程旅游信息技术(上海)有限公司 | Management device, system, method, electronic device and storage medium for thin client |
CN108009010A (en) * | 2017-12-01 | 2018-05-08 | 携程旅游信息技术(上海)有限公司 | The managing device of thin-client, system, method, electronic equipment, storage medium |
CN112106418A (en) * | 2018-03-20 | 2020-12-18 | 诺基亚技术有限公司 | Apparatus, method and computer program |
WO2019238092A1 (en) * | 2018-06-15 | 2019-12-19 | 华为技术有限公司 | Management method, device and system for virtual machine |
CN108958935A (en) * | 2018-06-28 | 2018-12-07 | 阿瑞思科技(成都)有限责任公司 | The method and its system of distributed computing are realized based on mobile terminal |
CN109040206A (en) * | 2018-07-09 | 2018-12-18 | 广东工业大学 | A kind of computing resource long-term allocation method based on mobile device cloud |
CN109040206B (en) * | 2018-07-09 | 2020-12-08 | 广东工业大学 | Computing resource long-term allocation method based on mobile device cloud |
CN112400301A (en) * | 2018-07-31 | 2021-02-23 | 韦斯特尔电子工业和贸易有限责任公司 | Method, apparatus, system and computer program for data distribution |
CN109246232A (en) * | 2018-09-30 | 2019-01-18 | 西安万像电子科技有限公司 | Data transmission method, system and device |
CN109766193A (en) * | 2019-01-29 | 2019-05-17 | 努比亚技术有限公司 | Calculating task executes method, server, terminal and computer readable storage medium |
CN110027596A (en) * | 2019-03-29 | 2019-07-19 | 北京交通大学 | A kind of Introduction of Train Operation Control System based on cloud computing |
CN110213180B (en) * | 2019-06-06 | 2022-04-05 | 腾讯科技(深圳)有限公司 | Network resource management method and device and cloud platform |
CN110213180A (en) * | 2019-06-06 | 2019-09-06 | 腾讯科技(深圳)有限公司 | Method for managing network resource, device and cloud platform |
CN112148439A (en) * | 2019-06-28 | 2020-12-29 | 浙江宇视科技有限公司 | Task processing method, device, equipment and storage medium |
CN112148439B (en) * | 2019-06-28 | 2024-03-08 | 浙江宇视科技有限公司 | Task processing method, device, equipment and storage medium |
CN110851529A (en) * | 2019-11-01 | 2020-02-28 | 腾讯科技(深圳)有限公司 | Computing power scheduling method and related equipment |
CN110851529B (en) * | 2019-11-01 | 2024-05-28 | 腾讯科技(深圳)有限公司 | Calculation power scheduling method and related equipment |
CN111461884A (en) * | 2020-03-31 | 2020-07-28 | 杭州溪塔科技有限公司 | Trusted computing service sharing method, device and system based on block chain |
CN111565101A (en) * | 2020-04-28 | 2020-08-21 | 上海物融智能科技有限公司 | Processing method and device for computing task |
CN114095577A (en) * | 2020-07-31 | 2022-02-25 | 中国移动通信有限公司研究院 | Resource request method and device, calculation network element node and calculation application equipment |
CN112068954A (en) * | 2020-08-18 | 2020-12-11 | 弥伦工业产品设计(上海)有限公司 | Method and system for scheduling network computing resources |
CN112274914A (en) * | 2020-10-29 | 2021-01-29 | 西安万像电子科技有限公司 | Method for realizing exclusive game, device for realizing exclusive game and cloud game system |
CN112416592A (en) * | 2020-11-26 | 2021-02-26 | 张新利 | Method for providing computing power of cloud server |
CN115378829A (en) * | 2021-05-17 | 2022-11-22 | 中国移动通信有限公司研究院 | Resource request method and device, computational power network element node and terminal equipment |
WO2022242342A1 (en) * | 2021-05-17 | 2022-11-24 | 中国移动通信有限公司研究院 | Resource request method and apparatus, computing power network element node, and terminal device |
Also Published As
Publication number | Publication date |
---|---|
CN103544064B (en) | 2018-03-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103544064A (en) | Cloud computing method, cloud management platform and client | |
US9015713B2 (en) | Facilitating customer-initiated virtual machine migration and swapping | |
Gupta | Above the clouds: a view of cloud computing | |
US8037187B2 (en) | Resource exchange management within a cloud computing environment | |
US20180095997A1 (en) | Secure automated resource-exchange system | |
US8706869B2 (en) | Distributed cloud placement software | |
US9129052B2 (en) | Metering resource usage in a cloud computing environment | |
CN114253793A (en) | Dynamic tracking control | |
US20170278100A1 (en) | Cryptographically assured zero-knowledge cloud service for composable atomic transactions | |
US11789767B2 (en) | Cooperative cloud infrastructure using blockchains for hardware ownership and access | |
Mulia et al. | Cloud workload characterization | |
US20120323821A1 (en) | Methods for billing for data storage in a tiered data storage system | |
CN111913875A (en) | Method and system for automatic tagging based on software execution tracking | |
Prabhakaran et al. | Cost-benefit Analysis of Public Clouds for offloading in-house HPC Jobs | |
US11010207B2 (en) | Metering software for cloud and non-cloud computer systems | |
US10554525B2 (en) | Tracking usage of computing resources | |
Marinescu et al. | An auction-driven self-organizing cloud delivery model | |
US20210042161A1 (en) | Scalable workflow engine with a stateless orchestrator | |
US10884818B2 (en) | Increasing processing capacity of virtual machines | |
CN109412865A (en) | A kind of virtual network resource allocation method, system and electronic equipment | |
Hu et al. | A dynamic User-Integrated cloud computing architecture | |
CN110633142B (en) | Block chain consensus method, management node, electronic device, and storage medium | |
TWM500311U (en) | The core cloud service system | |
US11763358B2 (en) | Per-tenant cost calculation in multi-tenant data storage system | |
US10922666B1 (en) | Resource management for logical and physical availability zones of a provider network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180313 Termination date: 20191028 |