CN104317633A - Cloud computing method facing high-performance scientific computation - Google Patents
Cloud computing method facing high-performance scientific computation Download PDFInfo
- Publication number
- CN104317633A CN104317633A CN201410515024.0A CN201410515024A CN104317633A CN 104317633 A CN104317633 A CN 104317633A CN 201410515024 A CN201410515024 A CN 201410515024A CN 104317633 A CN104317633 A CN 104317633A
- Authority
- CN
- China
- Prior art keywords
- user
- virtual machine
- machine instance
- resource
- performance
- 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.)
- Pending
Links
Abstract
The invention discloses a cloud computing method facing high-performance scientific computation. The cloud computing method comprises the following steps: a cloud portal carries out resource retrieval and distribution according to a user request, finds out corresponding data in shared storage and generates a virtual machine instance capable of directly accessing resources; the user accesses requested resources through a remote desktop of the virtual machine instance, pre-processes the resources, and stores a pre-processing result into the shared storage; the user calls a high-performance computation resource through the virtual machine instance; after the computation, the user checks a result through the remote desktop of the virtual machine instance, obtains the result of the high-performance computation through accessing the shared storage in the virtual machine instance and post-processes the result; the user takes a needed computation result back from a cloud environment through the remote desktop of the virtual machine instance. According to the cloud computing method facing the high-performance scientific computation, convenient pre-processing and post-processing of the scientific computation are provided for the user in the cloud environment; a convenient scientific computation service resource is provided for the user in the cloud environment through the high-performance computation resource and the shared storage.
Description
Technical field
The present invention relates to field of cloud calculation, particularly relate to a kind of cloud computing method towards high performance cementitious materisl.
Background technology
In recent years along with the explosive growth of science data, for science data investigation and application required for the storage that takies and computational resource get more and more.Cloud computing goes for large data to manage concentratedly, and provide a kind of new technique of relatively inexpensive service for user.But, be directed to computing power and require higher application, at present unripe be enough to meet big data quantity simultaneously storage, management and performance application achievement occur.The cloud computing environment of existing universality is more pay close attention to store and the service of popularity more, for the not well design of this intensive applications of scientific algorithm.
High-performance calculation resource plays very important role in quite a few scientific algorithm, and scientist needs the help of high-performance calculation to realize the analysis of science data.But high-performance calculation resource, building and operating and all need higher computer expertise and higher cost as cluster etc., can not accomplish to popularize.Therefore, scientific algorithm provides more cheap and easy acquisition high-performance calculation resource by certain mode is necessary for.Meanwhile, also need to carry out necessary pre-service and aftertreatment to adapt to the requirement of high-performance calculation and to get desirable result to data in scientific algorithm, this is equally also that the availability of scientific algorithm to computing environment proposes requirement.
But only by high-performance calculation resource, the method joined in cloud computing environment can not solve all contradiction.Because usually need in scientific algorithm to relate to very large data volume, and need to carry out carrying out aftertreatment after pre-service completes with the requirement or calculating that meet scientific algorithm to obtain the data meeting research and require to original science data.If in pretreated process, scientist needs that raw data is downloaded to this locality and processes, and then passes high in the clouds back; Or before aftertreatment, need that this locality is gone back in download complete for result of calculation and process; So, very large network flow consumption will certainly be produced, and very likely need the wait of long duration.Therefore, considering that the pre-service of scientific algorithm and last handling process are put into high in the clouds equally to be carried out, and provide corresponding use means for user, is a kind of extraordinary solution.
Inventor is realizing in process of the present invention, finds at least there is following shortcoming and defect in prior art:
1) due to not enough to the consideration of special applications in scientific algorithm, existing cloud environment can not meet the requirement of scientific algorithm;
2) due to technology and the cost barrier of high-performance calculation resource, high-performance calculation resource is not easily obtain or costliness for a lot of scientist;
Due to the singularity of data involved in scientific algorithm, the pre-service of science data and aftertreatment are needed to carry out enough considerations.
Summary of the invention
The invention provides a kind of cloud computing method towards high performance cementitious materisl, present invention accomplishes the requirement of scientific algorithm, achieve the convenience of high performance cementitious materisl, described below:
Towards a cloud computing method for high performance cementitious materisl, said method comprising the steps of:
User submits resource request by the resource retrieval interface at Yunmen family to Yunmen family;
Resource retrieval and distribution are carried out according to user's request in Yunmen family, in shared storage, find corresponding data, and generate can the virtual machine instance of direct access resources;
User has access to requested resource by the remote desktop of virtual machine instance and carries out pre-service to resource, by pre-processed results stored in shared storage;
User calls high-performance calculation resource by virtual machine instance;
High-performance calculation terminates, and user checks result by the remote desktop of virtual machine instance, and by accessing the result shared and store and obtain high-performance calculation in virtual machine instance, carries out aftertreatment to result; User fetches required result of calculation by the remote desktop of virtual machine instance from cloud environment.
Described user calls high-performance calculation resource by virtual machine instance and is specially:
User directly accesses high-performance calculation resource by virtual machine instance and calculates, and the centre of calculating and net result are stored into be shared in storage.
Described method also comprises:
User carries out individual cultivation by control desk to high-performance calculation resource, to meet computation requirement.
The beneficial effect of technical scheme provided by the invention is: the present invention by cloud computing environment to virtual machine, share and store and the integration of high-performance calculation resource, the high performance computing service in cloud computing environment is provided for scientific algorithm user.The present invention is pre-service and the aftertreatment that user provides scientific algorithm easily in cloud environment by virtual machine; By high-performance calculation resource with share and be stored as user provide scientific calculation service resource easily in cloud environment.
Accompanying drawing explanation
Fig. 1 is the overview flow chart of user operation in cloud computing environment;
Fig. 2 is the schematic diagram that in cloud computing environment, scientific algorithm pre-service and aftertreatment realize;
Fig. 3 is the mutual schematic diagram in cloud computing environment between various resource.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly, below embodiment of the present invention is described further in detail.
101: user submits resource request by the resource retrieval interface at Yunmen family to Yunmen family;
Wherein, these please comprise: existing data (optional) etc. in computing power, memory capacity and cloud computing environment.
102: resource retrieval and distribution are carried out according to user's request in Yunmen family, in shared storage, find corresponding data, and generate can the virtual machine instance of direct access resources;
First, Yunmen family is according to the resource request of user locating resource in shared storage; Then, Yunmen family creates a virtual machine instance in cloud environment, and enables this virtual machine instance directly access navigated to resource.Namely first user carries out Resource orientation by Yunmen family, and Resourse Distribute, according to the request of user to data (optional) existing in computing power, memory capacity and cloud computing environment, is carried out in Yunmen family; Corresponding data are found in shared storage.
103: user has access to oneself requested resource by the remote desktop of virtual machine instance and carries out pre-service to resource, by pre-processed results stored in shared storage;
Cloud computing environment, according to user's request dispatching storage space, computational resource and virtual machine instance, make virtual machine instance directly can access the storage space of corresponding data, computational resource and distribution, and this shared storage space only can be accessed by this user.User by virtual machine instance to request to science data carry out pre-service and by pre-processed results stored in shared storage, or directly upload data for scientific algorithm to sharing storage by virtual machine instance.As shown in Figure 2, user completes the pre-service of input data and consigns to high-performance calculation resource and calculate in cloud environment.
104: user calls high-performance calculation resource by virtual machine instance;
For needing the part using high-performance calculation resource in scientific algorithm, user directly accesses high-performance calculation resource by virtual machine instance and calculates, and the centre of calculating and net result are stored into be shared in storage.As shown in Figure 3, high-performance calculation resource directly can access shared storage to obtain data.
105: high-performance calculation terminates, user checks result by the remote desktop of virtual machine instance, and by accessing the result shared and store and obtain high-performance calculation in virtual machine instance, carries out aftertreatment to result; User fetches the result of calculation required for oneself by the remote desktop of virtual machine instance from cloud environment;
As shown in Figure 2, the result data of high-performance calculation is exported to user-defined aftertreatment by cloud environment, so that user obtains the data desired by oneself.
106: user carries out individual cultivation by control desk to high-performance calculation resource, to meet the computation requirement of oneself.
As shown in Figure 3, system can by specifying the authority of user, i.e. user authority management retrains user's calling and configuring for resource.
It will be appreciated by those skilled in the art that accompanying drawing is the schematic diagram of a preferred embodiment, the invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (3)
1. towards a cloud computing method for high performance cementitious materisl, it is characterized in that, said method comprising the steps of:
User submits resource request by the resource retrieval interface at Yunmen family to Yunmen family;
Resource retrieval and distribution are carried out according to user's request in Yunmen family, in shared storage, find corresponding data, and generate can the virtual machine instance of direct access resources;
User has access to requested resource by the remote desktop of virtual machine instance and carries out pre-service to resource, by pre-processed results stored in shared storage;
User calls high-performance calculation resource by virtual machine instance;
High-performance calculation terminates, and user checks result by the remote desktop of virtual machine instance, and by accessing the result shared and store and obtain high-performance calculation in virtual machine instance, carries out aftertreatment to result; User fetches required result of calculation by the remote desktop of virtual machine instance from cloud environment.
2. a kind of cloud computing method towards high performance cementitious materisl according to claim 1, is characterized in that, described user calls high-performance calculation resource by virtual machine instance and is specially:
User directly accesses high-performance calculation resource by virtual machine instance and calculates, and the centre of calculating and net result are stored into be shared in storage.
3. a kind of cloud computing method towards high performance cementitious materisl according to claim 1, it is characterized in that, described method also comprises:
User carries out individual cultivation by control desk to high-performance calculation resource, to meet computation requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410515024.0A CN104317633A (en) | 2014-09-29 | 2014-09-29 | Cloud computing method facing high-performance scientific computation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410515024.0A CN104317633A (en) | 2014-09-29 | 2014-09-29 | Cloud computing method facing high-performance scientific computation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104317633A true CN104317633A (en) | 2015-01-28 |
Family
ID=52372868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410515024.0A Pending CN104317633A (en) | 2014-09-29 | 2014-09-29 | Cloud computing method facing high-performance scientific computation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104317633A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105450668A (en) * | 2015-12-30 | 2016-03-30 | 中电长城网际系统应用有限公司 | Cloud security service implementing system and cloud security service implementing method |
| CN105915620A (en) * | 2016-05-05 | 2016-08-31 | 云神科技投资股份有限公司 | Data processing method based on cloud calculation network bearing balance |
| CN106250562A (en) * | 2016-08-24 | 2016-12-21 | 苏州蓝海彤翔系统科技有限公司 | Processing data information system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110258621A1 (en) * | 2010-04-14 | 2011-10-20 | International Business Machines Corporation | Autonomic Scaling Of Virtual Machines In A Cloud Computing Environment |
-
2014
- 2014-09-29 CN CN201410515024.0A patent/CN104317633A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110258621A1 (en) * | 2010-04-14 | 2011-10-20 | International Business Machines Corporation | Autonomic Scaling Of Virtual Machines In A Cloud Computing Environment |
Non-Patent Citations (1)
| Title |
|---|
| 张士勋: "《云计算在科学计算中的应用研究》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105450668A (en) * | 2015-12-30 | 2016-03-30 | 中电长城网际系统应用有限公司 | Cloud security service implementing system and cloud security service implementing method |
| CN105915620A (en) * | 2016-05-05 | 2016-08-31 | 云神科技投资股份有限公司 | Data processing method based on cloud calculation network bearing balance |
| CN106250562A (en) * | 2016-08-24 | 2016-12-21 | 苏州蓝海彤翔系统科技有限公司 | Processing data information system |
| CN106250562B (en) * | 2016-08-24 | 2019-08-23 | 苏州蓝海彤翔系统科技有限公司 | Processing data information system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109643312B (en) | Hosted query service | |
| CN105045607A (en) | Method for achieving uniform interface of multiple big data calculation frames | |
| US20130318084A1 (en) | Processing structured and unstructured data using offload processors | |
| US10248346B2 (en) | Modular architecture for extreme-scale distributed processing applications | |
| US9459897B2 (en) | System and method for providing data analysis service in cloud environment | |
| US20130080621A1 (en) | Hybrid storage devices | |
| CN102685246A (en) | Household design method and system based on cloud design | |
| US20160210077A1 (en) | Trans-cloud object based memory | |
| Chunlin et al. | Cost and energy aware service provisioning for mobile client in cloud computing environment | |
| WO2021074779A1 (en) | Providing direct data access between accelerators and storage in computing environment | |
| US20140047059A1 (en) | Method for improving mobile network performance via ad-hoc peer-to-peer request partitioning | |
| Arthi et al. | Energy aware cloud service provisioning approach for green computing environment | |
| CN103699656A (en) | GPU-based mass-multimedia-data-oriented MapReduce platform | |
| CN104021029A (en) | Spatial information cloud computing system and implementing method thereof | |
| CN104317633A (en) | Cloud computing method facing high-performance scientific computation | |
| WO2020215833A1 (en) | Offline cache method and apparatus, and terminal and readable storage medium | |
| KR102320324B1 (en) | Method for using heterogeneous hardware accelerator in kubernetes environment and apparatus using the same | |
| Nicolae et al. | Towards memory-optimized data shuffling patterns for big data analytics | |
| Liang et al. | Resource virtualization model using hybrid-graph representation and converging algorithm for cloud computing | |
| CN108319604B (en) | Optimization method for association of large and small tables in hive | |
| CN116860665A (en) | Address translation method executed by processor and related product | |
| CN105487929A (en) | Method for managing shared data of lens in cluster rendering process | |
| CN110120959A (en) | Big data method for pushing, device, system, equipment and readable storage medium storing program for executing | |
| Zhang et al. | Construction of cloud platform for personalized information services in digital library based on cloud computing data processing technology | |
| Zhang et al. | Speeding up vm startup by cooperative vm image caching |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150128 |
|
| RJ01 | Rejection of invention patent application after publication |