CN103179165A - Method and system of cloud computing and load balancing server - Google Patents
Method and system of cloud computing and load balancing server Download PDFInfo
- Publication number
- CN103179165A CN103179165A CN2011104392145A CN201110439214A CN103179165A CN 103179165 A CN103179165 A CN 103179165A CN 2011104392145 A CN2011104392145 A CN 2011104392145A CN 201110439214 A CN201110439214 A CN 201110439214A CN 103179165 A CN103179165 A CN 103179165A
- Authority
- CN
- China
- Prior art keywords
- computing
- electromagnetic material
- artificial electromagnetic
- process unit
- graphic process
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system of cloud computing formeta-materialsand a load balancing server. The method of cloud computing for meta-materials includes the steps of receiving the operation requests from a network server and data of meta-materials for operation; distributing at least a graphics processor computing node for the data of meta-materials for operation according to the operation requests; receiving the operation results of the data of meta-materials for operation from the graphics processor computing node, and sending the operation results of the data of meta-materials for operation from the graphics processor computing node to the network server. According to the above method, the method and system of cloud computing can solve the problem of slow matrix operation of personal computers, thereby decreasing the workload, improving the work efficiency and reducing the error rate.
Description
Technical field
The present invention relates to the cloud computing field, particularly relate to a kind of method for the artificial electromagnetic material cloud computing, system and load-balanced server.
Background technology
Artificial electromagnetic material is a kind of novel artificial material, and it utilizes substrate and be arranged on substrate and be periodically variable artificial micro-structural, realizes electromagnetic specific response, such as realizing electromagnetic focusing, parallel outgoing, dispersing etc.Each artificial micro-structural be to should there be specific geometry, and correspondingly has specific electromagnetic response parameter, and as refractive index, dielectric constant, magnetic permeability, resonance frequency etc., these electromagnetic response parameters have determined electromagnetic response characteristic just.The design artificial electromagnetic material, be exactly according to needed electromagnetic response feature, the parameters such as the refractive index on anti-each location point of release artificial electromagnetic material, dielectric constant, and then the parameter on each location point as goal condition, is designed the artificial micro-structural that has this goal condition on this location point respectively.
Make designed artificial micro-structural reach this goal condition, the computing that need to utilize sample taken at random by certain algorithm, constantly approach, live width, the line of namely constantly adjusting artificial micro-structural is thick, geometry is until till reaching goal condition.Therefore, even if the design of single artificial micro-structural, its amount of calculation is just very large; And simultaneously, the electromagnetic response of artificial electromagnetic material in order to obtain needs thousands of, tens thousand of artificial micro-structurals even up to a million jointly to respond usually, makes the amount of calculation of design of an artificial electromagnetic material very big.
Therefore, the rank of its complexity of a lot of designs of artificial electromagnetic material and large-scale data exceeds several orders of magnitude than other design.And use therein optimization method often is based on statistical algorithm, needs a large amount of matrix operations in optimizing process, but the computer of monomer is difficult to satisfy the requirement of such matrix computations, and a large amount of time spends on the wait operation result.
Therefore, need a kind of method, system and load-balanced server of the cloud computing for artificial electromagnetic material, carry out matrix operation problem slowly to solve active computer.
Summary of the invention
The technical problem that the present invention mainly solves is to provide a kind of method, system and load-balanced server of the cloud computing for artificial electromagnetic material, can utilize the cloud computing speed-up computation, increase work efficiency, reduce error rate, carry out matrix operation problem slowly to solve active computer.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is: a kind of method of the cloud computing for artificial electromagnetic material is provided, and the method comprises: receive from the computing request of the webserver and treat the data of the artificial electromagnetic material of computing; Be to treat at least one graphic process unit computing node of the data allocations of the artificial electromagnetic material of computing according to the computing request; The result of calculation of the artificial electromagnetic material data of computing is treated in reception from the graphic process unit computing node; Send the result of calculation of graphic process unit computing node to the webserver.
Wherein, receiving the step of result of calculation for the treatment of the artificial electromagnetic material data of computing from the graphic process unit computing node is specially: receive from the graphic process unit computing node and adopt the computing storehouse to treat the result of calculation of refractive index data, lattice dimensions and spacing of lattice after size, refractive index toroidal function or the discretization of the substrate material of the artificial electromagnetic material of computing, setting substrate.
Wherein, be that the step for the treatment of at least one graphic process unit computing node of artificial electromagnetic material data allocations of computing is specially according to the computing request: for the computing request distributes one or above graphic process unit computing node non-loaded or that load is lower.
Wherein, the step that sends the result of calculation of graphic process unit computing node to the webserver is specially: the form with file or chart sends the result of calculation of graphic process unit computing node to the webserver.
For solving the problems of the technologies described above, another technical solution used in the present invention is: a kind of load-balanced server is provided, this load-balanced server comprises: the first receiving element be used for to receive from the computing request of the webserver and treats the artificial electromagnetic material data of computing; Allocation units, being used for is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to the computing request; The second receiving element is used for receiving the result of calculation for the treatment of the artificial electromagnetic material data of computing from the graphic process unit computing node; Transmitting element is used for to the result of calculation of webserver transmission graphic process unit computing node.
For solving the problems of the technologies described above, another technical scheme that the present invention adopts is: the system that a kind of cloud computing for artificial electromagnetic material is provided, this system comprises: client terminal, in order to receive according to input command or script the artificial electromagnetic material data for the treatment of computing of uploading; The webserver in order to the artificial electromagnetic material data for the treatment of computing of reception from client terminal, and generates the computing request according to the artificial electromagnetic material data for the treatment of computing; Load-balanced server in order to receiving from the computing request of the webserver and to treat the artificial electromagnetic material data of computing, and is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to the computing request; The graphic process unit computing node calculates and result of calculation is returned to load-balanced server in order to the artificial electromagnetic material data for the treatment of computing;
Wherein, load-balanced server sends client terminal by the webserver with result of calculation.
Wherein, this system also comprises: artificial electromagnetic material data operation database server is used for when the artificial electromagnetic material data that the graphic process unit computing node is treated computing are calculated, for calculating provides the computing storehouse.
Wherein, load-balanced server comprises: the first receiving element be used for to receive from the computing request of the webserver and treats the artificial electromagnetic material data of computing; Allocation units, being used for is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to the computing request; The second receiving element be used for to receive the result of calculation for the treatment of refractive index data, lattice dimensions and spacing of lattice after size, refractive index toroidal function or the discretization of substrate material, setting substrate of the artificial electromagnetic material of computing from the graphic process unit computing node; Transmitting element is used for to the result of calculation of webserver transmission graphic process unit computing node.
Wherein, allocation units are further used for distributing one or above graphic process unit computing node non-loaded or that load is lower into the computing request.
Wherein, transmitting element is further used for sending the result of calculation of graphic process unit computing node to the webserver with the form of file or chart.
The invention has the beneficial effects as follows: the situation that is different from prior art, the present invention utilizes the method and system of cloud computing artificial electromagnetic material to be calculated to replace the calculating of personal computer with large-scale graphic process unit cluster, realized the extensive acceleration of calculating, to solve the existing artificial electromagnetic material related operation problem slowly of carrying out, avoid because the limited waste that can't complete at short notice the time and efforts that the computing requirement causes of computing capability, thereby minimizing workload, increase work efficiency, reduce error rate.
Description of drawings
Fig. 1 is the method flow diagram of the cloud computing that is used for artificial electromagnetic material of the embodiment of the present invention;
Fig. 2 is the particular flow sheet of step S2 of method of the cloud computing that is used for artificial electromagnetic material of the embodiment of the present invention;
Fig. 3 is the structural representation of the cloud computing system that is used for artificial electromagnetic material of the embodiment of the present invention;
Fig. 4 is the structural representation of load-balanced server of the cloud computing system that is used for artificial electromagnetic material of the embodiment of the present invention.
Embodiment
The present invention is described in detail below in conjunction with drawings and Examples.
See also Fig. 1, Fig. 1 is the method flow diagram of the cloud computing that is used for artificial electromagnetic material of the embodiment of the present invention.As shown in Figure 1, the method that is used for the cloud computing of artificial electromagnetic material comprises:
Step S1: receive from the computing request of the webserver and treat the artificial electromagnetic material data of computing;
In the present embodiment, the artificial electromagnetic material data include but not limited to refractive index data, lattice dimensions and the spacing of lattice after size, refractive index toroidal function or the discretization of substrate material, setting substrate of artificial electromagnetic material.
In the present embodiment, receiving the step of result of calculation for the treatment of the artificial electromagnetic material data of computing from the graphic process unit computing node is specially: receive from the graphic process unit computing node and adopt the computing storehouse to treat the result of calculation of refractive index data, lattice dimensions and spacing of lattice after size, refractive index toroidal function or the discretization of the substrate material of the artificial electromagnetic material of computing, setting substrate.
In the present embodiment, the webserver receives the computing request and treats that the mode of the artificial electromagnetic material data of computing also has different selections according to the difference of actual operating position.For example: then user's input message on computers imports the webserver into by http protocol, import information into the webserver by wireless telecommunications on wireless telecommunications system or when only supplying with inner the use user directly import information into the webserver by local area network (LAN).
Step S2: be to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to the computing request;
In the present embodiment, be that the step for the treatment of at least one graphic process unit computing node of artificial electromagnetic material data allocations of computing is specially according to the computing request: for the computing request distributes one or above graphic process unit computing node non-loaded or that load is lower.
See also Fig. 2, Fig. 2 is the particular flow sheet of step S2 of method of the cloud computing that is used for artificial electromagnetic material of the embodiment of the present invention.As shown in Figure 2, the concrete steps of step S2 comprise:
Step 401: receive the computation requests that the webserver imports into;
Step 402: if computation requests is comparatively complicated, the complex calculations request is split into the simple calculations request;
Step 403: judge whether to exist non-loaded graphic process unit computing node, if, enter step 404, if not, enter step 405;
Step 404: the computing request is assigned to non-loaded graphic process unit computing node;
Step 405: whether the load that judges each graphic process unit computing node is identical, if, enter step 406, if not, enter step 407;
Step 406: the graphic process unit computing node that the computing request is assigned to acquiescence;
Step 407: the computing request is assigned to the lower graphic process unit computing node of load.
Step S3: receive the result of calculation for the treatment of the artificial electromagnetic material data of computing from the graphic process unit computing node;
In the present embodiment, adopt high performance graphic process unit computing node, than existing methods, effectively having reduced high-performance troops to the demand of power consumption, higher performance and power dissipation ratio have been realized, and the graphic process unit computing node is having greater advantage aspect computing linear algebra and figure, and the design of artificial electromagnetic material is having larger demand aspect computing linear algebra and figure, so computing node adopts graphic process unit to come computing.
In other embodiments, require also can change into the computing node of other types or adopt the mode that makes up with better adaptation computing request according to the computing in the design of concrete artificial electromagnetic material.
Step S4: the result of calculation that sends the graphic process unit computing node to the webserver.
In the present embodiment, the step that sends the result of calculation of graphic process unit computing node to the webserver is specially: the form with file or chart sends the result of calculation of graphic process unit computing node to the webserver.
In the present embodiment, to obtain a refractive index as the artificial electromagnetic material structure of X as example, the result of calculation of returning is the information such as refractive index of arrangement mode and each cellular construction of the cellular construction in each lamella, arrangement mode is specifically used JPEG (joint photographic experts group) (JPG), bitmap (BMP), the graphical formats such as streaming network graphical format (PNG) represent, also can represent with industrialized standard drawing swap file (dxf) and gerber files (gerber) when demand of industrial production is arranged.
In other embodiments, in fact the form that sends can be also diversified, the user can decide mode and the form of transmission according to actual conditions and the hobby of self, the user can select directly to send to result on the log-on webpage of client by the webserver, also can select result is directly sent to by the webserver mail mailbox of user's appointment, even can set in advance the path that is stored in local computer, automatically store according to the path after result of calculation draws.
See also Fig. 3, Fig. 3 is the structural representation of the cloud computing system that is used for artificial electromagnetic material of the embodiment of the present invention, as shown in Figure 3, the cloud computing system 200 that is used for artificial electromagnetic material comprises: client terminal 210, the webserver 220, load-balanced server 230, graphic process unit computing node 240 and operational data storehouse server 250.
Wherein, client terminal 210 is in order to receive according to input command or script the artificial electromagnetic material data for the treatment of computing of uploading; The webserver 220 is in order to receiving the artificial electromagnetic material data for the treatment of computing from client terminal, and generates the computing request according to the artificial electromagnetic material data for the treatment of computing; Load-balanced server 230 is in order to receiving from the computing request of the webserver and to treat the artificial electromagnetic material data of computing, and is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to the computing request; Graphic process unit computing node 240 calculates and result of calculation is returned to load-balanced server in order to the artificial electromagnetic material data for the treatment of computing; Budget database server 250 is used for when the artificial electromagnetic material data that graphic process unit computing node 240 is treated computing are calculated, for calculating provides the computing storehouse.
Wherein, load-balanced server 230 sends client terminal 210 by the webserver 220 with result of calculation.
In the present embodiment, client terminal 210 further comprises the management function of user right and user's space.Client terminal 210 in the present embodiment refers to the page operational outfit that uses webpage portal form, and described client provides a visual image processor operation to submit the interface to the user, can realize cross-platform requirement simultaneously.The management in integrated user right and space in webpage portal has reduced the management difficulty of cloud computing.The user only need to login the pass just can use whole functions, and need to be when not using other functions login account repeatedly, and webpage portal also offers user's space of user, the user can store a certain amount of data or data inside with user-friendly, the user's space the inside also includes the function setting of portal page, and the user can inside revise according to hobby.
In the present embodiment, provide user-defined function by webpage portal, the user can set according to the hobby of oneself.For example: select the upload file form of acquiescence, select the color of the page and title.
In the present embodiment, also provide interpretative function by webpage portal, the user can become some data by this portal page the language of oneself wanting with data translation, the page itself also can change according to user's needs the language of the page or directly the language on the page just be translated.
In other embodiments, client terminal can also be ios, android, linux and mac.
See also Fig. 4, Fig. 4 is the structural representation of load-balanced server of the cloud computing system that is used for artificial electromagnetic material of the embodiment of the present invention.As shown in Figure 4, load balancing service 230 comprises: the first receiving element 2301, allocation units 2302, the second receiving element 2303 and transmitting element 2304.
Wherein, the first receiving element 2301 is used for receiving from the computing request of the webserver and treats the artificial electromagnetic material data of computing; It is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to the computing request that allocation units 2302 are used for; The second receiving element 2303 is used for receiving the result of calculation for the treatment of the artificial electromagnetic material data of computing from the graphic process unit computing node; Transmitting element 2304 is used for sending to the webserver result of calculation of graphic process unit computing node.
In the present embodiment, the second receiving element 2303 is used for receiving the result of calculation for the treatment of refractive index data, lattice dimensions and spacing of lattice after size, refractive index toroidal function or the discretization of substrate material, setting substrate of the artificial electromagnetic material of computing from the graphic process unit computing node.
In the present embodiment, what the second receiving element 2303 received treats the result of calculation JPG specifically of the artificial electromagnetic material data of computing from the graphic process unit computing node, BMP, PNG etc. have indicated the distribution of cellular construction in lamella and the graphic file of arrangement in the artificial electromagnetic material, and the checkout result that receives when demand of industrial production is arranged is with industrialized standard dxf and gerber file.
In the present embodiment, load-balanced server 230 first splits into a plurality of simple calculations requests with a complex calculations request, then distributes load to try one's best the load of each graphic process unit computing node 240 of balance in the hope of reaching the highest utilization ratio according to the loading condition of each graphic process unit computing node 240.
How load-balanced server 230 specifically distributes load to see also above description about a kind of step S2 of method of the cloud computing for artificial electromagnetic material, repeats no more herein.
In other embodiments, a webserver can connect a plurality of different load-balanced servers with the different types of computing request of correspondence, and the webserver sends to them on different load-balanced servers according to the type difference of computing request.
in the present embodiment, adopt high performance graphic process unit computing node, than existing methods, effectively having reduced high-performance troops to the demand of power consumption, realized that higher performance and power dissipation ratio and graphic process unit computing node are having greater advantage aspect computing linear algebra and figure, and in the designing and calculating of artificial electromagnetic material, linear algebra and figure aspect also had larger demand, so the computing node of the present embodiment adopts the graphic process unit computing node, but also can change the computing node of other types into or adopt the mode that makes up to adapt to better the computing requirement according to concrete actual conditions.
In the present embodiment, all kinds of algorithms of design that graphic process unit computing node 240 is integrated, can satisfy the demand of the artificial electromagnetic material design of most of types, for example: particle cluster algorithm, in the situation that given training data, spread one or more targets of a little approaching object space by utilizing particle cluster algorithm in artificial electromagnetic material structural parameters spaces.Then seek out corresponding cellular construction based on optimal value, integrate with substrate data more at last and namely get result.The user can add as required voluntarily, deletion or replace Algorithm, in order to can satisfy dissimilar computing requirement.
In the present embodiment, operational data storehouse server 250 the is integrated contour performance computing of CUBLAS, CUFFT storehouse provides the computing storehouse for graphic process unit computing node 240 calculates.
See also above description for the method for the cloud computing that is used for artificial electromagnetic material about the concrete workflow of the cloud computing system that is used for artificial electromagnetic material and operation principle, repeat no more herein.
in sum, be different from the situation of prior art, the calculating that whole process migration to the high performance cloud computing platform that the present invention utilizes the method and system of cloud computing that large-scale artificial electromagnetic material is designed carries out artificial electromagnetic material with large-scale graphic process unit cluster replaces the calculating of personal computer, realized the extensive acceleration of calculating, carry out artificial electromagnetic material related operation problem slowly to solve existing personal computer, avoided because the limited waste that can't complete at short notice the time and efforts that the computing requirement causes of computing capability, thereby minimizing workload, increase work efficiency, reduce error rate.
The above is only embodiments of the invention; not thereby limit the scope of the claims of the present invention; every equivalent structure or equivalent flow process conversion that utilizes specification of the present invention and accompanying drawing content to do; or directly or indirectly be used in other relevant technical fields, all in like manner be included in scope of patent protection of the present invention.
Claims (10)
1. the method for a cloud computing, is characterized in that, described method comprises:
Reception is from the computing request of the webserver and treat the artificial electromagnetic material data of computing;
Be to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to described computing request;
The result of calculation of the artificial electromagnetic material data of computing is treated in reception from described graphic process unit computing node;
Send the result of calculation of the artificial electromagnetic material data of described graphic process unit computing node to the webserver.
2. method according to claim 1, it is characterized in that, the artificial electromagnetic material data of computing are treated in described reception from the graphic process unit computing node the step of result of calculation is specially: receive from described graphic process unit computing node and adopt the computing storehouse to treat the result of calculation of refractive index data, lattice dimensions and spacing of lattice after size, refractive index toroidal function or the discretization of the substrate material of the artificial electromagnetic material of computing, setting substrate.
3. method according to claim 1, it is characterized in that, described is that the step for the treatment of at least one graphic process unit computing node of artificial electromagnetic material data allocations of computing is specially according to the computing request: for described computing request distributes one or above graphic process unit computing node non-loaded or that load is lower.
4. method according to claim 1, it is characterized in that, the described step that sends the result of calculation of graphic process unit computing node to the webserver is specially: the form with file or chart sends the result of calculation of described graphic process unit computing node to the webserver.
5. a load-balanced server that is used for artificial electromagnetic material, is characterized in that, comprising:
The first receiving element be used for to receive from the computing request of the webserver and treats the artificial electromagnetic material data of computing;
Allocation units, being used for is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to described computing request;
The second receiving element is used for receiving the result of calculation for the treatment of the artificial electromagnetic material data of computing from described graphic process unit computing node;
Transmitting element is used for to the result of calculation of the described graphic process unit computing node of webserver transmission.
6. a system that is used for the cloud computing of artificial electromagnetic material, is characterized in that, comprising:
Client terminal is in order to receive according to input command or script the artificial electromagnetic material data for the treatment of computing of uploading;
The webserver in order to the described artificial electromagnetic material data for the treatment of computing of reception from client terminal, and treats that according to described the artificial electromagnetic material data of computing generate the computing request;
Load-balanced server in order to receiving from the computing request of the described webserver and to treat the artificial electromagnetic material data of computing, and is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to described computing request;
The graphic process unit computing node is in order to treat that to described the artificial electromagnetic material data of computing calculate and described result of calculation is returned to load-balanced server;
Wherein, described load-balanced server sends result of calculation by the network server to client terminal.
7. system according to claim 6, is characterized in that, described system also comprises:
Artificial electromagnetic material data operation database server is used for when the artificial electromagnetic material data that the graphic process unit computing node is treated computing are calculated, for described calculating provides the computing storehouse.
8. system according to claim 6, is characterized in that, described load-balanced server comprises:
The first receiving element be used for to receive from the computing request of the webserver and treats the artificial electromagnetic material data of computing;
Allocation units, being used for is to treat at least one graphic process unit computing node of the artificial electromagnetic material data allocations of computing according to described computing request;
The second receiving element be used for to receive the result of calculation for the treatment of refractive index data, lattice dimensions and spacing of lattice after size, refractive index toroidal function or the discretization of substrate material, setting substrate of the artificial electromagnetic material of computing from described graphic process unit computing node;
Transmitting element is used for to the result of calculation of the described graphic process unit computing node of webserver transmission.
9. system according to claim 8, is characterized in that, described allocation units are further used for distributing one or above graphic process unit computing node non-loaded or that load is lower into described computing request.
10. system according to claim 8, is characterized in that, described transmitting element is further used for sending the result of calculation of described graphic process unit computing node to the webserver with the form of file or chart.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110439214.5A CN103179165B (en) | 2011-12-26 | The method of cloud computing, system and load-balanced server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110439214.5A CN103179165B (en) | 2011-12-26 | The method of cloud computing, system and load-balanced server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103179165A true CN103179165A (en) | 2013-06-26 |
| CN103179165B CN103179165B (en) | 2016-12-14 |
Family
ID=
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106326007A (en) * | 2016-08-22 | 2017-01-11 | 北京信安世纪科技有限公司 | Distribution operation method, device and system |
| CN106651741A (en) * | 2016-12-09 | 2017-05-10 | 西华大学 | Cloud computing-based graphic processing system |
| CN108694151A (en) * | 2017-04-09 | 2018-10-23 | 英特尔公司 | Computing cluster in universal graphics processing unit is seized |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101596113A (en) * | 2008-06-06 | 2009-12-09 | 中国科学院过程工程研究所 | A kind of CT parallel reconstructing system and formation method |
| US20100312995A1 (en) * | 2009-06-09 | 2010-12-09 | Zillians Incorporated | Virtual world simulation systems and methods utilizing parallel coprocessors, and computer program products thereof |
| CN102232282A (en) * | 2010-10-29 | 2011-11-02 | 华为技术有限公司 | Method and apparatus for realizing load balance of resources in data center |
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101596113A (en) * | 2008-06-06 | 2009-12-09 | 中国科学院过程工程研究所 | A kind of CT parallel reconstructing system and formation method |
| US20100312995A1 (en) * | 2009-06-09 | 2010-12-09 | Zillians Incorporated | Virtual world simulation systems and methods utilizing parallel coprocessors, and computer program products thereof |
| CN102232282A (en) * | 2010-10-29 | 2011-11-02 | 华为技术有限公司 | Method and apparatus for realizing load balance of resources in data center |
Non-Patent Citations (3)
| Title |
|---|
| 卢风顺等: "CPU/GPU协同并行计算研究综述", 《CPU/GPU协同并行计算研究综述》 * |
| 张庆科等: "基于异构模式的云计算关键技术研究", 《山东科学》 * |
| 赖麟: "色散光子晶体的GPU-FDTD并行算法研究", 《中国优秀硕士学位论文全文数据库基础科学辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106326007A (en) * | 2016-08-22 | 2017-01-11 | 北京信安世纪科技有限公司 | Distribution operation method, device and system |
| CN106651741A (en) * | 2016-12-09 | 2017-05-10 | 西华大学 | Cloud computing-based graphic processing system |
| CN106651741B (en) * | 2016-12-09 | 2019-12-13 | 西华大学 | graphics processing system based on cloud computing |
| CN108694151A (en) * | 2017-04-09 | 2018-10-23 | 英特尔公司 | Computing cluster in universal graphics processing unit is seized |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qi et al. | A QoS-aware virtual machine scheduling method for energy conservation in cloud-based cyber-physical systems | |
| Satpathy et al. | Crow search based virtual machine placement strategy in cloud data centers with live migration | |
| Abd Elaziz et al. | IoT workflow scheduling using intelligent arithmetic optimization algorithm in fog computing | |
| Zhang et al. | Joint edge server placement and service placement in mobile-edge computing | |
| Islambouli et al. | Optimized 3D deployment of UAV-mounted cloudlets to support latency-sensitive services in IoT networks | |
| CN103237037A (en) | Media format conversion method and system based on cloud computing architecture | |
| KR20160040516A (en) | Power balancing to increase workload density and improve energy efficiency | |
| CN104915407A (en) | Resource scheduling method under Hadoop-based multi-job environment | |
| CN111597043B (en) | Full scene edge calculation method, device and system | |
| Ghetas | A multi-objective Monarch Butterfly Algorithm for virtual machine placement in cloud computing | |
| CN102307133A (en) | Virtual machine scheduling method for public cloud platform | |
| Pham et al. | Applying Ant Colony System algorithm in multi-objective resource allocation for virtual services | |
| CN103281374A (en) | Method for rapid data scheduling in cloud storage | |
| CN104731528A (en) | Construction method and system for storage service of cloud computing block | |
| CN103988179A (en) | Optimization mechanisms for latency reduction and elasticity improvement in geographically distributed datacenters | |
| Li et al. | Scalable and dynamic replica consistency maintenance for edge-cloud system | |
| Gokuldhev et al. | Local pollination-based moth search algorithm for task-scheduling heterogeneous cloud environment | |
| Tyagi et al. | Computing resource optimization of big data in optical cloud radio access networked industrial Internet of Things | |
| Li et al. | Data analytics for fog computing by distributed online learning with asynchronous update | |
| Huang et al. | Computation offloading for multimedia workflows with deadline constraints in cloudlet-based mobile cloud | |
| Chen et al. | Resource utilization and cost optimization oriented container placement for edge computing in industrial internet | |
| CN106874108B (en) | Technology for minimizing number of micro clouds in mobile cloud computing | |
| CN103179167A (en) | Method and system for cloud computing and load balancing server | |
| Wang | Collaborative task offloading strategy of UAV cluster using improved genetic algorithm in mobile edge computing | |
| CN113014649A (en) | Cloud Internet of things load balancing method, device and equipment based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |