CN1829215A - Drive storage application system based on network and gridding - Google Patents
Drive storage application system based on network and gridding Download PDFInfo
- Publication number
- CN1829215A CN1829215A CN 200510020449 CN200510020449A CN1829215A CN 1829215 A CN1829215 A CN 1829215A CN 200510020449 CN200510020449 CN 200510020449 CN 200510020449 A CN200510020449 A CN 200510020449A CN 1829215 A CN1829215 A CN 1829215A
- Authority
- CN
- China
- Prior art keywords
- resource
- storage
- grid
- network based
- network
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention refers to driver storage application system Based on network and lattice storage. It contains data transmission sharing protocol (NDFP GDFP protocol) based on network and lattice, contains communication mode between application system and to realize uniform virtual store resource sharing, driver storing installation module, network and lattice based on storage technology (NDS / GDS technology) supporting several iso- subsystem resource and storage application equipment mapping and defining, physics storage resource and computing resource management module, physics storage resource and computing resource distributing storage and coordination by plurality of storage application equipment distributed different place, solving plurality of load balance problem. Said invention can change large scale complicate system into single virtual system to reach saving cost, raising efficiency and resource sharing.
Description
Technical field
The present invention relates to areas of information technology, particularly a kind of drive storage application system of Network Based and grid.
Background technology
Network storage equipment system provides the application system of data message carrier, and being mainly used in provides data centralization, density data access, mass data, real time data distribution, data management, exchanges data, the contour performance service of Information Security by network to the outside in the network storage environment.No matter be based at present the network storage equipment system that the optical-fibre channel technical standard also is based on up-to-date IP memory technology standard, as its biggest advantage of application system such as data warehouse system, DSS, group system, standby system, tolerant systems is exactly the interface that understandable management storage systems easily is provided for the user, has solved the user to being the storage application demand of core with traditional server, the network equipment.But these traditional network storage equipment systems all can't will be distributed in the virtual resource system that the resource such as computational resource, storage resources, data resource, information resources, knowledge resource of some isomerized sub-systems of this locality or wide scope is formed in the mode of overall sharing in the parallel distributed mode, and can be according to Resource Availability, function, performance, cost and quality of services for users demand, the various resources that share, selection and polymerization are distributed in some isomerized sub-system management domains.Simultaneously also there is the unbalanced problem of load between the low and a plurality of network storage equipments of physics utilization ratio of storage resources in traditional network storage equipment system.
Summary of the invention
Purpose of the present invention just is to provide a kind of drive storage application system of Network Based and grid, solve existing network memory technology and network storage equipment system and can't satisfy the requirement of high performance computing system the network storage equipment, particularly the multiple tracks concurrent job problem of the I/O performance requirement during the access mass data is simultaneously unified the virtual resource system that virtual memory forms in the parallel distributed mode with the resource of some isomerized sub-systems of this locality or wide scope.And can have linear retractable according to the demand of high performance computing system at capacity and aspect of performance, collaborative work in parallel storage networking and grid environment.To overcome the shortcoming of above-mentioned existing legacy network storage system.
For achieving the above object, the drive storage application system of Network Based and grid comprises:
Transfer of data Network Based and grid is shared agreement (NDFP/GDFP agreement)
Transfer of data Network Based and grid is shared agreement (NDFP/GDFP agreement), mainly be the communication mode that is used to define between the drive storage application system of some isomerized sub-systems and Network Based and grid, and how realize realizing the resource overall sharing with the virtual resource system that the form of unifying virtual memory forms.
Module is installed in the driver storage
Module is installed in driver storage, mainly is to be used to support to be distributed in the mapping relations that the physical store resource of some isomerized sub-systems of this locality or wide scope and computational resource are implemented to the mapping of storage application apparatus with Network Based and driver memory technology (NDS/GDS technology) grid and define both.
Physical store resource and managing computing resources module
Physical store resource and managing computing resources module, mainly be to be used for physical store resource and computational resource being done unified memory allocated and coordinated management, solve the load balance problem between a plurality of storage application apparatus system and realize the resource overall sharing by being distributed in different local a plurality of Network Based and storage application apparatuss grid.
The storage application apparatus of Network Based and grid
The storage application apparatus of Network Based and grid, mainly be to be used for storing the magnanimity physical store resource of the some isomerized sub-systems that are distributed in this locality or wide scope and computational resource, data resource, information resources, knowledge resource etc. in the parallel distributed mode, form a virtual memory resource system by unified virtual memory, build " huge resource system network library " collaborative work in parallel storage networking and grid environment simultaneously.
The present invention is by the drive storage application system of Network Based and grid, having realized being distributed in the physical store resource of some isomerized sub-systems of this locality or wide scope and computational resource unifies virtual memory in the parallel distributed mode and forms a virtual resource system, thereby large complicated system is formed a virtual single storage system, save cost, raise the efficiency, reach the overall sharing of resource, also can provide the storage application service for the high performance computing system based on grid computing of Future Development.
Description of drawings
Fig. 1 is a building-block of logic of the present invention.
In Fig. 1 in Network Based and the grid storage application apparatus place of tamping represent to have had resource.
Embodiment
As shown in Figure 1, the drive storage application system of this Network Based and grid provided by the invention, the transfer of data that comprises Network Based and grid are shared the storage application apparatus that agreement (NDFP/GDFP agreement), driver storage are installed module, physical store resource and managing computing resources module, Network Based and grid.This drive storage application system Network Based and grid is shared agreement (NDFP/GDFP agreement) by network or grid application environment with the transfer of data of Network Based and grid provides high-performance data storage share service to outside high performance computing system or some isomerized sub-system users.Outside high performance computing system or some isomerized sub-system users install module by the driver storage and share agreement (NDFP/GDFP agreement) with the transfer of data of Network Based and grid and visit to this storage application system, and this storage application system is with data and the information person that directly returns to the external demand.
In the drive storage application system of Network Based and grid, communication mode between the drive storage application system of the mainly responsible some isomerized sub-systems of definition of the shared agreement (NDFP/GDFP agreement) of transfer of data Network Based and grid and Network Based and grid, in the time will setting up communication between some isomerized sub-systems in outside and the storage application system, the shared agreement (NDFP/GDFP agreement) of transfer of data with grid Network Based just erects both communication according to definition mode, and the communication between the two of guaranteeing can correctly be carried out.In addition, the transfer of data of Network Based and grid is shared agreement (NDFP/GDFP agreement), and also will to guarantee to store the application apparatus system be that mode with parallel distributed realizes that the physical store resource and the computational resource of some isomerized sub-systems of this locality or wide scope can unify virtual resource system of virtual memory formation, the overall sharing of realization resource.
After setting up correct communication between the drive storage application system of externally some isomerized sub-systems and Network Based and grid, the storage of the driver of application system is installed module and is responsible for the request of data of the some isomerized sub-systems in outside is forwarded to corresponding stored application apparatus system.The driver storage is installed module and has been located the mapping relations of data resource with storage application apparatus system, pass through data resource of being located and the mapping relations of storing the application apparatus system, the storage application apparatus position that module can be determined the request msg place of outside some isomerized sub-systems is installed in the driver storage, thereby exactly the request of data of the some isomerized sub-systems in outside is forwarded to corresponding stored application apparatus system with Network Based and driver memory technology (NDS/GDS technology) grid.According to the external data request outside high performance computing system or some isomerized sub-system user resources are done unified distribution storage and coordinated management by physical store resource and managing computing resources module then, unify its various resources of virtual memory by Network Based and storage application apparatus grid in the parallel distributed mode, and form a virtual storage system by unified virtual memory.Here physical store resource and managing computing resources module also can be according to the resource and the loading conditions of formed single virtual storage system, by carrying out cooperation, rationally solve the load balance problem between a plurality of storage application apparatus system with Network Based and storage application apparatus system grid.
Drive storage application system by Network Based and grid provided by the invention, the physical store resource and the computational resource that can will be distributed in some isomerized sub-systems of this locality or wide scope by driver memory technology (NDS/GDS technology) realization with Network Based and grid are unified virtual resource system of virtual memory formation in the parallel distributed mode, thereby large complicated system is formed a virtual single storage system, and be different local by being distributed in this virtual single storage system space, the some storage application apparatuss system of different field constitutes.For effectively utilizing the space of the unified virtual storage system that is constituted by these some storage application apparatuss systems, the physical store resource will be carried out the storage and uniform distribution and the coordinated management of resource to being distributed in different local storage application apparatus systems with the managing computing resources module, when certain storage application apparatus system needs newly-built virtual memory resource, physical store resource and managing computing resources module will be according to the distribution condition of resource, provide the storage idling-resource to this storage application apparatus system, like this for then effectively solve the load balance problem between a plurality of storage application apparatus system by the formed single storage system of unified virtual memory.
The drive storage application system of Network Based and grid has just formed a virtual single storage system with large complicated system after the driver memory technology (NDS/GDS technology) by Network Based and grid realizes that the physical store resource of some isomerized sub-systems and computational resource are unified virtual memory and formed a virtual resource system in the parallel distributed mode.This virtual single storage system provides one " huge resource system network library " for the user who is dispersed in diverse geographic location, each user can participate in becoming " node " of this " huge resource system network library " as required by the unified memory interface request that drives, whole like this " resource system network library " will be made up of thousands of " node ", becomes increasing.Physical store resource in the drive storage application system of Network Based and grid and managing computing resources module can do that storage and uniform is distributed and coordinated management to the various information in this " huge resource system network library ", and share agreement according to the transfer of data of Network Based and grid the definition mode of resource-sharing is made comprehensive resource-sharing, thereby eliminated the resource isolated island.
Network Based and drive storage application system grid provided by the invention can be realized being distributed in the physical store resource of some isomerized sub-systems of this locality or wide scope and computational resource and unify virtual memory in the parallel distributed mode and form a virtual resource system, changed the mode that people share data message, thereby large complicated system can be formed a virtual single storage system, save cost, raise the efficiency, reach the overall sharing of resource, eliminated the resource isolated island.This Network Based and drive storage application system grid has the following advantages:
1, can provide the storage I/O ability that exceeds several magnitude such as modern diverse network memory apparatus system, polymerization storage throughput will reach per second hundreds of GB and count to TB, rather than several MB of per second, thereby satisfy the requirement of high performance computing system, particularly the multiple tracks concurrent job I/O performance requirement during the access mass data simultaneously to memory device.
2, along with the continuous expansion of physical store resource and computational resource scale, the continuous growth of using, this Network Based and drive storage application system grid can have linear retractable at capacity and aspect of performance, the capacity of a reflection can reach tens petabytes (PB), Chinese mugwort byte (EB) even Yao's byte at most, and this storage application system itself can collaborative work in parallel storage networking and grid environment.
3, according to the isomerism that is distributed in the plurality of sub system resource of this locality or wide scope, this Network Based and drive storage application system of grid can be arranged on different local, but can keep single reflection, realization physical store resource and computational resource are unified virtual memory in the parallel distributed mode and are formed a virtual resource system, reach the resource overall sharing.
4, drive storage application system Network Based and grid can solve the unbalanced problem of load between the low and a plurality of network storage equipment of the physical store resource utilization system that the existing network memory apparatus system exists effectively.
Claims (7)
1, a kind of drive storage application system of Network Based and grid comprises:
The transfer of data of Network Based and grid is shared agreement (NDFP/GDFP agreement), a virtual resource system realization resource overall sharing that is used to define the communication mode between the drive storage application system of some isomerized sub-systems and Network Based and grid and realizes forming with the form of unifying virtual memory;
Module is installed in driver storage, is used to support to be distributed in the mapping relations that the physical store resource of some isomerized sub-systems of this locality or wide scope and computational resource are implemented to the mapping of storage application apparatus with Network Based and driver memory technology (NDS/GDS technology) grid and define both;
Physical store resource and managing computing resources module, be used for physical store resource and computational resource being done unified memory allocated and coordinated management, solve the load balance problem between a plurality of storage application apparatus system and realize the resource overall sharing by being distributed in different local a plurality of Network Based and storage application apparatuss grid;
The storage application apparatus of Network Based and grid, be used for storing the magnanimity physical store resource of the some isomerized sub-systems that are distributed in this locality or wide scope and computational resource, data resource, information resources, knowledge resource etc. in the parallel distributed mode, form a virtual memory resource system by unified virtual memory, build " huge resource system network library " collaborative work in parallel storage networking and grid environment simultaneously.
2, by the described system of claim 1, it is characterized in that the shared agreement (NDFP/GDFP agreement) of described transfer of data Network Based and grid, comprise communication mode and virtual resource between application system.
3, by the described system of claim 1, it is characterized in that the storage of described driver installs module, comprise local and some isomerized sub-system resources of wide scope and store the mapping of application apparatus and the driver memory technology (NDS/GDS technology) of the Network Based and grid of definition.
4, by the described system of claim 1, it is characterized in that described physical store resource and managing computing resources module, comprise resource allocation and coordination and load balance problem.
5,, it is characterized in that described Network Based and storage application apparatus grid comprises unified virtual memory formed " huge resource system network library " by the described system of claim 1.
6,, it is characterized in that described " huge resource system network library " is to unify the virtual resource system that virtual memory forms by the physical store resource and the computational resource of the same alike result of some isomerized sub-systems by the described system of claim 5.
7, by the described system of claim 1, unified driver memory interface and shared " huge resource system network library " that the drive storage application system that it is characterized in that also comprising whole Network Based and grid provides to some isomerized sub-systems and user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510020449 CN1829215A (en) | 2005-03-04 | 2005-03-04 | Drive storage application system based on network and gridding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200510020449 CN1829215A (en) | 2005-03-04 | 2005-03-04 | Drive storage application system based on network and gridding |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1829215A true CN1829215A (en) | 2006-09-06 |
Family
ID=36947343
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200510020449 Pending CN1829215A (en) | 2005-03-04 | 2005-03-04 | Drive storage application system based on network and gridding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1829215A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110119303A (en) * | 2013-11-12 | 2019-08-13 | 微软技术许可有限责任公司 | Construct virtual mainboard and virtual memory facilities |
| US10701159B2 (en) | 2013-12-17 | 2020-06-30 | Huawei Technologies Co., Ltd. | Method and apparatus for implementing device sharing |
-
2005
- 2005-03-04 CN CN 200510020449 patent/CN1829215A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110119303A (en) * | 2013-11-12 | 2019-08-13 | 微软技术许可有限责任公司 | Construct virtual mainboard and virtual memory facilities |
| US10701159B2 (en) | 2013-12-17 | 2020-06-30 | Huawei Technologies Co., Ltd. | Method and apparatus for implementing device sharing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107066319B (en) | Multi-dimensional scheduling system for heterogeneous resources | |
| Amokrane et al. | Greenhead: Virtual data center embedding across distributed infrastructures | |
| Alicherry et al. | Network aware resource allocation in distributed clouds | |
| US7809769B2 (en) | Database partitioning by virtual partitions | |
| Tsai et al. | Two-tier multi-tenancy scaling and load balancing | |
| CN102209087A (en) | Method and system for MapReduce data transmission in data center having SAN | |
| CN103914399A (en) | Disk cache method and device in parallel computer system | |
| CN104735095A (en) | Method and device for job scheduling of cloud computing platform | |
| CN113504954A (en) | Method, system and medium for calling CSI LVM plug-in, dynamic persistent volume provisioning | |
| CN103095788A (en) | Cloud resource scheduling policy based on network topology | |
| US20210132845A1 (en) | Method for storage management, electronic device and computer program product | |
| Souri et al. | A survey for replica placement techniques in data grid environment | |
| US20210120077A1 (en) | Multi-tenant isolated data regions for collaborative platform architectures | |
| CN106528265A (en) | Virtual machine management method and system based on distributed SAN (Storage Area Network) storage architecture | |
| CN109302494A (en) | A kind of configuration method of network store system, device, equipment and medium | |
| CN101014044A (en) | Network GIS system and data transmitting method thereof | |
| Albrecht et al. | Making work queue cluster-friendly for data intensive scientific applications | |
| US20040093390A1 (en) | Connected memory management | |
| CN1829215A (en) | Drive storage application system based on network and gridding | |
| CN104580422A (en) | Cluster rendering node data access method based on shared cache | |
| CN1327198A (en) | Method system for sectional processing environment isomer customer service device and program product | |
| CN1613243A (en) | Method, apparatus and system for routing messages within a packet operating system | |
| Yu et al. | Towards predictable performance via two-layer bandwidth allocation in cloud datacenter | |
| CN1474275A (en) | System of intellignent network storage device based on virtual storage | |
| Brimberg et al. | Allocation of queuing facilities using a minimax criterion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |