CN101079736A - Modeled network resource positioning method - Google Patents
Modeled network resource positioning method Download PDFInfo
- Publication number
- CN101079736A CN101079736A CN 200710100325 CN200710100325A CN101079736A CN 101079736 A CN101079736 A CN 101079736A CN 200710100325 CN200710100325 CN 200710100325 CN 200710100325 A CN200710100325 A CN 200710100325A CN 101079736 A CN101079736 A CN 101079736A
- Authority
- CN
- China
- Prior art keywords
- resource
- node
- gridding
- grid
- user
- 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
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000013507 mapping Methods 0.000 claims abstract description 36
- 230000008569 process Effects 0.000 claims description 43
- 230000000295 complement effect Effects 0.000 claims description 9
- 230000004807 localization Effects 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 6
- 238000012423 maintenance Methods 0.000 claims description 5
- 238000012217 deletion Methods 0.000 claims description 4
- 230000037430 deletion Effects 0.000 claims description 4
- 230000000977 initiatory effect Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims 2
- 235000012364 Peperomia pellucida Nutrition 0.000 claims 1
- 240000007711 Peperomia pellucida Species 0.000 claims 1
- 230000008676 import Effects 0.000 claims 1
- 239000007787 solid Substances 0.000 abstract 4
- 230000006870 function Effects 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 8
- 239000000284 extract Substances 0.000 description 7
- 230000015654 memory Effects 0.000 description 7
- 238000012856 packing Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 4
- 238000004435 EPR spectroscopy Methods 0.000 description 3
- 229920000181 Ethylene propylene rubber Polymers 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 229940028444 muse Drugs 0.000 description 2
- GMVPRGQOIOIIMI-DWKJAMRDSA-N prostaglandin E1 Chemical compound CCCCC[C@H](O)\C=C\[C@H]1[C@H](O)CC(=O)[C@@H]1CCCCCCC(O)=O GMVPRGQOIOIIMI-DWKJAMRDSA-N 0.000 description 2
- 206010026749 Mania Diseases 0.000 description 1
- 240000000233 Melia azedarach Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a modeling grid resource position method in the grid resource position technique field in the network system, which is characterized by the following: converting the resource ergodic cues regarded the corresponding logical solid in the resource module of the user needed grid resource as the root to a resource position entrance of the grid resource based on the grid resource module comprised by the logical solid and the modeling relation, the mapping relation of the logical solid and the grid resource, the mapping relation of the modeling relation and the grid resource attribute in the module; replacing the resource position cues of the logical solid as the root node; acquiring the resource position cues registered by the grid node from the corresponding registered node by the user node; acquiring the positioned grid resource from the corresponding grid node; replacing the non-root node in the ergodic cues as the corresponding grid resource attribute. The invention provides the good expandable property and the good flexibility, which simplifies the consistent maintaining problem of the registered information on the registered node.
Description
Technical field
Modeled gridding resource localization method belongs to distributed computing technology and system field, relates in particular to high-performance grid computing technology field
Background technology
People find the solution the field of problem in continuous expansion in the modern scientific research, the problem that is run into also becomes increasingly complex, and scale is increasing, addresses these problems needed computing capability and is also increasing substantially, and local computational resource can't satisfy the demands.Fast development along with computer and network technologies, a lot of tissues and R﹠D institution all have the very strong supercomputer of computing capability, but these machines are often because just isolated not playing one's part to the full for our unit serves is in idle condition in the most of the time.Therefore, break the restriction of region, the collaborative all kinds of resources that extensively distribute on network of using have become necessary requirement.
In order to satisfy above-mentioned this requirement, its target is that cross-region realizes sharing and collaborative work of heterogeneous resource, eliminates information island and resource isolated island just for the proposition of grid and development.The technology relevant with it comprises network technology, XML technology, Web service technology (web service), high-performance calculation etc.Maturation gradually along with grid, combine Open Grid Service framework (the Open Grid Services Architecture of service-oriented theory, OGSA) become grid circle de facto standards, and the Web service resource framework (WS-Resource Framework WSRF) is the up-to-date realization standard of OGSA.A grid system (being also referred to as grid environment) of deferring to the OGSA framework generally includes following three class entities:
● shared resource.The object of computing capability, energy independent operating really is provided, as the computer of operating system has been installed, network, storage systems etc. are the material bases that grid system is rely and moved.Shared resource in the grid externally shows as autonomous unit individuality by individual/tissue (being called " Resource Owner " or " keeper ") contribution.The Resource Owner can be used to limit the use of grid system to its shared resource to the qualification such as " only be no more than in machine loading and could be used by grid system " in addition of the shared resource of its contribution at 50% o'clock.The shared resource that adds grid system is also referred to as grid node usually, each grid node in grid environment, all have the unique sign of the overall situation (such as, use the IP address to identify grid node).
● mesh services.Based on the local runtime environment (as operating system) of shared resource, for reaching certain functional purpose, and the program code of developing according to specific mesh standard/standard (as WSRF).Mesh services is deployed on the grid node, and the computing capability of use shared resource is come the request of process user.The process that the function that the software that is equipped with on the shared resource, existing program etc. is had is issued as mesh services is called the service packing.A plurality of mesh services can be made up according to certain flow process, be issued as the service of workflow grid, provide more strong functions to the user.Workflow service and general mesh services are as broad as long for the user, are the mesh services that has met the specifiable lattice standard/specification.
● the user.Call the entity of mesh services, be called user node again.
The user obtains the computing capability of shared resource contribution in the grid environment by the visit mesh services.The mesh services that realizes according to the WSRF standard whenever receives an access request, all can generate Service Source in real time or select a Service Source that instantiation is good in advance to handle request this time; The feature that user's request produces in processing procedure then shows by the attribute of this Service Source.Such as, a certain grid node deploy has a batch processing service, after it receives a batch processing task requests of user's submission, is responsible for the execution matters of this user task with generating a batch processing Service Source; The batch processing Service Source has " task status " this Resource Properties, and the value of this attribute (as " in the formation ", " carrying out ", " end " etc.) has then illustrated respective user task executions situation.Therefore, Service Source has reacted shared resource to the situation that the user provides computing power, claims that such Service Source is a gridding resource.Specifically, gridding resource is a logical concept on shared resource, and they provide the function corresponding with mesh services by the computing capability of virtual and polymerization shared resource to the user.The reference address of a gridding resource (being locating information) all is that the overall situation is unique in whole grid environment.
Be in correspondence with each other between mesh services and the gridding resource.The gridding resource of their correspondences of different mesh services is also different.According to the WSRF standard, this corresponding relation has " generally " and " Singleton " two kinds of patterns.The mesh services of Singleton pattern only can corresponding gridding resource at run duration, and all requests are finished by this gridding resource; The mesh services of general modfel can corresponding a plurality of gridding resources at run duration, and these resources can be asked the back dynamic instanceization receiving the user, also can just generate in advance when initialization.In general, the Singleton pattern makes and is used less, and general modfel is comparatively general.In view of the difference on the mesh services function, the gridding resource of their correspondences is also very big in the difference of aspects such as structure, content, realization, and this phenomenon is called the isomerism of gridding resource.
For the mesh services that realizes by general modfel, how to manage its corresponding gridding resource, determine by the Resource Owner fully.When grid node has certain limitation to simultaneously treated number of tasks, generally can adopt the way of " generating one group of grid resource in advance selects for mesh services ", limit the number of user's request by the number of gridding resource.Also " generating in advance " and " dynamic instanceization " these two kinds of ways can be mixed and use: at first generate one group of grid resource in order to using; When user's request comes, if there is idle gridding resource just therefrom to choose one, gridding resource returned idle condition after request disposed; If there is not idle gridding resource, the gridding resource that dynamic instanceization is new, gridding resource will be destroyed after request disposed.No matter any mode, gridding resource on the grid node all is dynamic changes: the one, and the value of Resource Properties changes along with the execution of user's request, the 2nd, gridding resource can generate in real time along with the arrival of user request, along with user's processing of request finishes and withers away.
When the user capture grid system, at first need to know the information of grid node in the system, mesh services and gridding resource, could judge that can this grid system satisfy his computation requirement.For reaching this purpose, selected usually some grid node is a register node, is responsible for all or part of grid node in the maintenance system, mesh services, and the information of gridding resource.Register node is a notion in logic, and the user can be known the access method of register node.By register node, the user can inquire any information that this register node is safeguarded.In the grid system one or more register node can be arranged, each entity that need register can arrive on any register node to be registered, the content of registration is called the metadata of registering entities, such as application software, reference address and the agreement etc. of cpu type, operating system, current state, disk size, deployment.Metadata is the data of data of description, is used to provide the structure and the meaning of data, is an abstract concept.The metadata has here been showed the content of aspects such as the state of registering entities, feature and function in the grid system to the user.
In sum, in the grid environment of deferring to OGSA framework and WSRF standard, autonomous grid node according to certain standard and agreement local externally resources shared according to its resource model service packing of determining, and dynamically generate or use the good gridding resource of initialization in advance according to actual conditions.Gridding resource on the different grid nodes all may be different on syntax and semantics (structure and content), presents isomerism, and their organization and administration mode also may be different.The user must position gridding resource earlier before using the computing capability that gridding resource presented.
Present gridding resource targeting scheme adopts the form of " level registration-direct index " mostly, as the Monitoring and Discovery System (MDS) among the Globus Toolkit, register node can be constructed as arbitrary levels, can register mutually between a plurality of register node, even constitute the index relative of ring-type; The information of gridding resource can successively arrive other register node successively by its register node of directly registering.This " level registration-direct index " though mode have favorable tissue, its target is a gridding resource, granularity is refinement too, mainly has following shortcoming:
● when system scale increased, the number of gridding resource is also corresponding to be increased, and being in the log-on message that high-rise register node need safeguard in hierarchical structure sharply increases, and causes single point failure, poor expandability easily;
● owing to the dynamic of gridding resource, need to guarantee the consistency of the information of preserving on gridding resource self and a plurality of register node.When the information change of gridding resource is frequent, consistency maintenance will produce a large amount of expenses;
● because register node all is isomorphism, this requires in the system gridding resource on all grid nodes all must realize very flexible according to identical standard/standard (or even identical instrument).
Under the background of " level registration-direct index ", modeled gridding resource localization method is based on the characteristic of grid node autonomy, the target of register node is turned to static relatively resource model from the dynamic grid resource on the grid node, loosened the restriction on grid node realization gridding resource institute code requirement/instrument, both increase flexibility, simplified the consistency maintenance and the scalability problem of log-on message on the register node again.
Summary of the invention
The object of the present invention is to provide a kind of modeled gridding resource localization method, so that for realizing that the resource-sharing under the grid environment lays the foundation.This method is a starting point with the autonomy of grid node, goes up isomery and dynamic gridding resource by related its of static relatively grid node resource model, and gridding resource that will be corresponding with this resource model is located clue and is registered to register node as metadata.Because the nature static of grid node resource model adopts this method can effectively solve by the caused log-on message consistency maintenance of the dynamic of gridding resource problem, can improve extensibility simultaneously; On the other hand, grid node can adopt the mode of liking arbitrarily that local shared resources (LSR) is carried out the mesh services packing, has both embodied the autonomy of grid node, has strengthened the flexibility that improves system again.
For content of the present invention is described better, at first carry out following explanation and explanation of nouns:
1. resource model.Grid node externally shows as autonomous unit individuality, but its inside may include a plurality of son individualities.Resource model is that grid node is carried out result after abstract, and mainly comprise relation (Relations) two parts between logic entity (LogicEntities) and logic entity: logic entity is represented the inherent element of grid node; Contact between two inherent elements of relation expression is also referred to as the modeling relation, has one to one (1: 1), and (1: N), (N: N) etc. quantity is described attribute to multi-to-multi to one-to-many.Such as, can with the grid node of group type abstract be " internal node " and " process " these two elements; Process operates on the internal node, so between these two elements " operating in " this contact is arranged; Can move a plurality of processes on the internal node, and a process can only be on an internal node, so be the relation of one-to-many between " inner node " and " process "; Therefore, the resource model of modeling comprises " internal node " and " process " these two logic entities in a manner described, and " operating in (one-to-many) " this modeling relation.For the sake of simplicity, the quantity of modeling relation is described attribute and is only mentioned just now on the ground of necessity, and other all omits.For the resource model that comprises N logic entity, mathematical definition is as follows:
ResourceModel＝{LogicEntities，Relations}
LogicEntites＝{LogicEntity
i|i＝1，2，...，N}
Relations＝{R(i，j)＝<log?icEntity
i，log?icEntity
j>|i≠j，i∈{1，...，N}，j∈{1，...，N}}
R(i，j)(1∶N-1)＝R(j，i)(N-1∶1)
2. the resource model of Lian Tonging.The process of grid node being carried out modeling is artificial, thereby it is different often to carry out abstract result according to various objectives.Such as, from the angle that distributed system is formed, a group of planes can be modeled as and comprise " internal node ", " process " and " inter-cluster network " these three logic entities, and the resource model of " operating in " this relation.In a resource model, if having direct relation between two logic entities, maybe can by and one or more other logic entities between relation come indirect association, claim these two logic entities to be communicated with.If all can be communicated with between any two logic entities, then claim this resource model to be communicated with.An establishment mould that is communicated with two logic entities concerns that sequence is called the paths between these two logic entities, as＜R (i, k
1), R (k
2, k
3) ..., R (k
a, j)〉and be LogicEntity
iAnd LogicEntity
jBetween a paths, wherein, i ∈ 1 ..., N}, j ∈ 1 ..., N}, k
x∈ 1 ..., N}, i ≠ j, k
x≠ i, k
x≠ j, and k
x≠ k
y, x, y=1,2 ..., a, a≤N-2.For top example, because have only the relation of " operating in ", can only get in touch " internal node " and " process " these two logic entities, and " inter-cluster network " exists with it doesn't matter between other logic entities, be in isolated state, so this resource model is disconnected.
3. the resource of the traversal clue of resource model and grid node is located clue.Resource location clue is to define at having the grid node that is communicated with resource model.Serve in the process of packing at grid node, each logic entity in the resource model all is embodied as a mesh services, and the modeling relation is embodied as an attribute of gridding resource.Be in correspondence with each other between mesh services and the gridding resource, so also have mapping relations between gridding resource and the logic entity; Each modeling relation that logic entity participates in all will be an attribute on the gridding resource corresponding with this logic entity.
[1] resource model can regard with the logic entity to be the summit as, and it is the figure on limit that modeling is closed, thereby can have the tree of generation.In the resource model of a connection, can arrive other any logic entities from logic entity LE; According to the thought that generates tree, it be that root, modeling close with LE is the tree of non-root node that the set of paths that LE is arrived other any logic entities is organized as one, is called a traversal clue of resource model;
[2] all gridding resources on the grid node are all corresponding with a logic entity in the resource model, thereby from any one gridding resource GR, the sequence by the gridding resource attribute corresponding to the modeling relation constitutes can arrive other arbitrary mess resources;
[3] logic entity of supposing the GR correspondence is LE, is the traversal clue of root for one with LE, root node LE is replaced to the locating information of GR, and non-root node (being the modeling relation) replaces with corresponding gridding resource attribute, and the tree that claims to replace after finishing is a resource location clue of grid node.
[4] because may a more than paths between two logic entities, so be that the traversal clue of root also has many with same logic entity.Accordingly, be that the resource location clue of root also has a plurality of with same gridding resource.
4. the index path of gridding resource.In resource model, a logic entity can arrive another logic entity by a paths, changes the relation of the modeling in this paths into corresponding gridding resource attribute successively, then can obtain an index path between gridding resource.Suppose＜R (LE
K1, LE
K2), R (LE
K2, LE
K3) ..., R (LE
K (i-1), LE
Ki) be from logical resource LE in the resource model
KiTo logical resource LE
iA paths, with R (LE
i, LE
j) corresponding gridding resource attribute is RP (LE
i, LE
j), then corresponding with this paths index path is＜RP (LE
K1, LE
K2), RP (LE
K2, LE
K3) ..., RP (LE
K (i-1), LE
Ki).
5. each gridding resource can both come unique identification by its reference address (being locating information) in grid environment, so also can directly replace with gridding resource in the locating information of the local gridding resource that can not cause ambiguity.
Principal character of the present invention is: under the support of OGSA framework and WSRF standard, grid node adopts the thought towards model, make up the metadata of registering to register node according to its resource model, make user node to locate clue, and come in view of the above the dynamic grid resource on the grid node is carried out Search and Orientation by the resource of inquiring about grid node to register node.At first grid node is carried out Resource Modeling, guarantee that its resource model is the logic entity relational network of a connection; Determine the mapping relations of logic entity and gridding resource, modeling relation and gridding resource attribute, and realize in view of the above with the initialization grid node on mesh services and gridding resource; Choosing certain gridding resource arbitrarily is resource location inlet, constructs a resource location clue according to the traversal clue in the resource model, and its metadata as grid node is registered to register node.User node is understood on the grid node after the contact between gridding resource by resource model, according to resource location clue, can effectively locate the dynamic grid resource on the grid node from resource location inlet.
Based on above-mentioned feature, the summary of the invention of modeled gridding resource localization method comprises:
1. the Resource Modeling of grid node and serviceization packing realize.The Resource Owner of grid node or keeper at first carry out modeling to local physical resource, guarantee the connectivity of its resource model, and the mapping relations of definite logic entity and gridding resource, modeling relation and gridding resource attribute, grid node is embodied as the gridding resource relational network corresponding with resource model.Specifically comprise:
1) provide the initial resource model InitRM (having N logic entity) of grid node according to actual conditions, judge whether this model is communicated with, if not, then carry out following operation:
[1] InitRM being regarded as with the logic entity is the summit, and it is the figure on limit that modeling is closed, and finds out disjoint connected subgraph in the resource model, makes InitRM=P1 ∪ P2 ∪ ... ∪ Pm, m≤n
[2] respectively from each connected subgraph Pi (i=1 ..., select a logic entity LE arbitrarily in m)
Pi
[3] in InitRM, add a complementary logic entity LE
N+1With m complementary modeling relation＜LE
N+1, LE
Pi, i=1 ..., m
Attention: the purpose of step [3] is to make between m disjoint connected subgraph to interconnect, and the simplest method directly is exactly to add the individual modeling relation of m-1; Also can be to add k complementary logic entity LE
N+1..., LE
N+kWith m+k-1 complementary modeling relation.Choose which kind of way and depend on the Resource Owner fully.InitRM has added the resource model FCRM that becomes a connection after complementary logic entity and the modeling relation.
2) mapping relations of logic entity and gridding resource, modeling relation and gridding resource attribute.The service packing of grid node is carried out based on FCRM, each logic entity among the FCRM all is embodied as a mesh services, and each modeling relation is embodied as an attribute of gridding resource.Because mesh services and gridding resource are in correspondence with each other, so also be mutual mapping between gridding resource and the logic entity, modeling between logic entity relation also is used to represent the contact between gridding resource; Each modeling relation that logic entity participates in all will be to be mapped as a attribute on the gridding resource of this logic entity with all.Specifically comprise:
[1] for complementary logic entity, the logic entity among the promptly non-initial resource model InitRM is embodied as the mesh services of Singleton pattern, has a unique gridding resource; And the logic entity among the InitRM can be embodied as the mesh services of general modfel or Singleton pattern.
[2] if logic entity LE
iAnd LE
jBetween exist modeling concern R (i, j), R (i, j) corresponding Resource Properties be RP (i, j), corresponding LE no matter then
iOr corresponding LE
jGridding resource all have Resource Properties RP (i, j);
[3] RP (i, value j) is a list type, each is all being stored has modeling with this gridding resource and concerns R (i, the reference address of other gridding resources j) (being locating information);
[4] if logic entity LE
iAnd LE
jBetween modeling to close be R (i, j) (1:N), then corresponding LE
iGridding resource on RP (i, value j) has only one, and corresponding LE
jGridding resource on R (i, value j) can have multinomial.
3) the XML language representation is used in the mapping of mapping, modeling relation and the gridding resource attribute of connection resource model, logic entity and the gridding resource of grid node, and be kept at grid node this locality with the form of file.
2. the object that relates in the gridding resource position fixing process logically is divided into register node, grid node and user node, overall relation between the three and interaction diagrams are as shown in Figures 1 and 2.Wherein,
1) user node is for initiating the grid entity of gridding resource location.
2) grid node provides computing capability by gridding resource to user node.As shown in Figure 3, grid node has a real-time storage space, and its module that is used for the gridding resource location comprises:
■ initialization service module InitResources.The initialization operation that is used for grid node mainly is responsible for the initial work of grid node running environment, and workflow is as follows:
[1] in the real-time storage space, loads the mapping of resource model, logic entity and the gridding resource preserved with the form of XML file and the mapping of modeling relation and gridding resource attribute.Describe for convenient, represent the content of this XML file with ModelDescription.
[2] according to ModelDescription, and the Resource Owner shares the setting (maximum number of the number of initialized all kinds of gridding resources, all kinds of gridding resources etc. in advance when starting as grid node of aspect to local computing capability, usually preserve these settings with configuration file), instantiation mesh services and gridding resource are preserved in the real-time storage space their information in the mode of XML.
[3] make up resource location clue.May have many communication paths between two logic entities in the resource model, thereby also have many index paths accordingly between two gridding resources, and then the resource on grid node location clue also can have many, needs selected wherein one as the registration metadata.Operating procedure is as follows:
A) gridding resource GR of picked at random enters the mouth as the resource location from the gridding resource of instantiation;
B), find out the logic entity LE of GR correspondence in resource model according to the mapping between logic entity and the gridding resource;
C) the traversal clue TGlue that to select a LE at random in resource model be root;
D) root of TGlue is replaced with the locating information of GR, and, the non-root node among the TGlue is replaced with corresponding gridding resource attribute according to the mapping of modeling relation and gridding resource attribute.Like this, just having obtained one is the resource location clue of resource location inlet with GR.
[4] resource being located clue registers to register node as the log-on message of grid node.
■ model element data service module ModelMetadata.From acting on, it provides the Resource Modeling of grid node and the readable explanation of gridding resource implementation with the form of XML to the user, i.e. the XML of the contents such as mapping of mapping of grid node resource model, logic entity and gridding resource and modeling relation and gridding resource attribute describes.The method that this module provides comprises:
Resource model output interface XMLObject GetResourceModel ()
Function: called by user node, be used to obtain the resource model of grid node
Input: do not have
Output: the XML that returns resource model (comprising logic entity and modeling relation) to user node describes
Operation: visit real-time storage space, the XML that extracts resource model among the ModelDescription describes part
Mapping relations output interface XMLObject GetModelMapping ()
Function: called by user node, be used to obtain the implementation of gridding resource on the grid node, be i.e. the mapping of mapping of logic entity and gridding resource and modeling relation and gridding resource attribute
Input: do not have
Output: return the mapping relations of describing with XML to the user, comprise the mapping of logic entity and gridding resource and the mapping of modeling relation and gridding resource attribute
Operation: in the visit real-time storage space, obtain the part that is used to describe above-mentioned mapping among the ModelDescription
Content description output interface XMLObject GetAll ()
Function: called by user node, be used to obtain the ModelDescription of grid node
Input: do not have
Output: return the ModelDescription that explains with the XML form to the user
Operation: in the visit real-time storage space, obtain ModelDescription
The direct inquiry service module of ■ RelationService.The user has the locating information of the gridding resource of specific modeling relation by this module inquiry and specifiable lattice resource.The query interface that provides is:
XMLObject?Query(?)
Function: called by user node, return and specify gridding resource to have to specify the locating information of the gridding resource of modeling relation
Input: specify by user node, comprise
● a gridding resource (initResource) locating information, with string representation;
● a gridding resource attribute relationship, this gridding resource attribute must be able to correspond to a modeling relation, and are that initResource has, with string representation.
Output: return the gridding resource locating information of representing with list type to the user, each in the tabulation all is that the XML of a gridding resource locating information describes
Operation: visit the real-time storage space, from the information of current initResource, extract the value of this Resource Properties of relationship
■ positioning service module LocationService.Complicated gridding resource positioning service is provided, and the user can obtain and specify gridding resource to have the locating information of the gridding resource that appointment directly or indirectly gets in touch by this module.The locating interface that provides is:
XMLObject?Query(?)
Function: called by user node, return and specify gridding resource to have the locating information of the gridding resource that appointment directly or indirectly gets in touch
Input: specify by user node, comprise
● the locating information of a gridding resource (GR), with string representation;
● index path LocationChain=＜RP (LE of a gridding resource
K1, LE
K2), RP (LE
K2, LE
K3) ..., RP (LE
Ki, LE
Kj);
● one group of filtering rule filterRule, whether appointment is returned with the locating information that GR has the gridding resource of certain modeling relation.
Output: return the gridding resource locating information of representing with list type to the user, each in the tabulation all is the gridding resource locating information of describing with XML.
Operation:
(a) set and need be Results to the gridding resource locating information tabulation that the user return, other has two tabulation Temp1 and Temp2 to be used for interim storage, with Results, be changed to sky, GR is added Temp1;
(b) if Temp1 is empty, directly return Results; Otherwise:
I. take out first gridding resource attribute RP of LocationChain, afterwards RP is deleted from LocationChain;
Ii. for the gridding resource crg among each Temp1,, call RelationService, each the gridding resource locating information that comprises among its RP is all added Temp2 if crg possesses this Resource Properties of RP;
The locating information of these gridding resources to be returned if iii. specify among the filterRule, then the element among the Temp2 is copied to Results;
(c) if also have the gridding resource attribute among the LocationChain, empty Temp1 earlier, all elements of Temp2 be copied to Temp1, empty Temp2 afterwards, repeat step (b),
(d) if there has not been remaining gridding resource attribute among the LocationChain, return Results to the user.
3) register node provides the registration/cancellation service and the inquiry service of resource location clue.As shown in Figure 4, it disposes real-time storage space and these two storage areas of back up memory space, and comprises as lower module:
■ registration service module RegisterService is used to register the resource location clue of grid node, and the registration interface that provides is:
boolear?register(?)
Function: called by grid node, be used to register the resource location clue of grid node
Input: provide by grid node, comprise
● the sign gridSiteID of grid node;
● the resource location clue LG of grid node gridSiteID;
Output: return the Boolean whether expression succeeds in registration to grid node
Operation: check that at first whether the real-time storage space has preserved this resource location clue of LG of gridSiteID, if do not have, then adds gridSiteID and LG in the real-time storage space to the form of XML.If operate successfully, return true, otherwise return false.
■ nullifies the resource location clue that service module UnregisterService is used to nullify grid node, and addressable cancellation interface is:
boolean?unregister(?)
Function: called by grid node, be used to nullify the resource location clue of grid node
Input: provide by grid node, comprise
● the sign gridSiteID of grid node;
● the resource location clue LG of grid node gridSiteID;
Output: return the Boolean that success or not is nullified in expression to grid node
Operation: the resource location clue LG of deletion gridSiteID from the real-time storage space.If delete successfully, return true, otherwise return false.If the resource of this gridSiteID location clue LG not just originally in the real-time storage space does not then need to carry out deletion action, directly return true.
■ index service module I ndexService provides the index service of grid node resource location clue, and the interface of searching that provides is:
XMLObject?find()
Function: called by user node, be used to inquire about the resource location clue of grid node
Input: provide the sign gridSite of target gridding node by user node
Output: the form with tabulation is located clues to all resources of registering of user node gridSite on this register node, and each resource location clue is all represented with XML
Operation: retrieval real-time storage space, if preserve the resource location clue of one or more gridSite, the form that then they is organized into tabulation is returned to user node
3. the gridding resource on the grid node has dynamic characters, and the prerequisite that they can be positioned is that grid node is finished initialization, enters normal operating condition.The user at first needed to obtain grid node and its resource location clue before the locating resource, this means that the initialization of register node is finished and registered/operate as normal of retrieval service.
1) workflow of grid node integral body comprises initialization and operational process two parts of grid node as shown in Figure 5.
The initialization of ■ grid node is finished by initialization service InitResource, and grid node enters normal operating conditions afterwards.
During ■ grid node operate as normal, whether monitoring constantly has comes from the request of user node about the resource location, if having, carries out following processing operation:
[1] input parameter is extracted in reception and judgement user request;
[2] if the request of model element data, services ModelMetadata, then describe to the XML that user node returns grid node resource model and/or mapping relations, change [5] by invoked interface (GetResourceModel, GetModelMapping or GetAll);
[3] if the directly request of inquiry service RelationService, then return according to input parameter and specify gridding resource to have to specify the modeling relation all gridding resources locating information---this process is finished by the query interface of RelationService, change [5];
[4] if the request of positioning service LocationService, then by from request input, extracting the initial gridding resource of inquiry, resource index clue and filtering rule as a result, calling the RelationService interface inquires about successively, and think user's return results according to filtering rule---this process is finished by the locating interface of LocationService, changes [5];
[5] after the user returns request results, get back to wait state.
2) workflow of register node integral body is seen accompanying drawing 5, comprises initialization, asks to intercept and handle and the data backup part.
The initialized key step of ■ register node
[1] the reset running environment of register node;
[2], load grid node and the resource thereof registered the front and locate information such as clue according to the log-on message backup file in the back up memory space;
[3] start request and intercept handling process and data backup flow process.
The ■ request is intercepted handling process and is started afterwards constantly intercept whether the request that comes from user node or grid node is arranged, if having, carries out following operation:
[1] input parameter is extracted in reception and judgement user request;
[2] if the request of registration service RegisterService at first judges whether to be repeated registration, if not, then being saved in the real-time storage space---this process is finished by its register interface;
[3] if nullify the request of service UnregisterService, from the real-time storage space deletion by the required parameter appointment log-on message---this process is finished by its unregister interface;
[4] if the request of index service IndexService, then according to the input parameter appointment index key retrieval real-time storage space---this process is finished by its find interface;
[5] return the result of request to the user, get back to wait state.
■ data backup flow process is the flow process of safeguarding of log-on message, intercept the handling process parallel running with request, regularly the log-on message in the real-time storage space (being the resource location clue of grid node) is saved in back up memory space, carries out the loading of log-on message when being convenient to register node initialization next time.Its workflow is:
[1] timer of log-on data backup;
[2] judge whether to be backup time, if, then from the real-time storage space, extract all log-on messages, be saved to back up memory space with the form of XML file, this XML file will cover the last backup file of preserving;
[3] wait for the arrival of backup time next time.
4. the positioning action of gridding resource is initiated by user node.User node at first obtains the resource location clue of selected grid node to register node inquiry, then by and selected grid node between one or many mutual, obtain on the selected grid node locating information of all its gridding resource of being concerned about.Idiographic flow as shown in Figure 7, key step has:
[1] user node is judged the resource location clue that self whether has target gridding node TGS, if do not have, the index service of visit register node is obtained the resource location clue GRLG of TGS;
[2] user node judges whether self understands the resource model of TGS, if not, the model element data, services by TGS is then obtained in the following content of representing with XML one or multinomial: the mapping between the mapping between XML description, gridding resource and the logic entity of resource model, gridding resource attribute and modeling relation.Like this, user node can complete understanding TGS be gone up the contact between all gridding resources;
[3],, from GRLG, extract the index path LPath between the inlet Entry of GR and resource location for any one gridding resource GR that the user was concerned about on this grid node based on the understanding of TGS being gone up the contact between all gridding resources;
[4] if there is direct modeling relation between GR and the Entry, promptly only comprise an element among the LPath, then with Entry and LPath as required parameter, send request to the direct inquiry service of TGS, change [6];
[5] if be not direct modeling relation between GR and the Entry, filtering rule filterRule as a result is set then, Entry, LPath and filterRule as required parameter, are sent request to the positioning service of TGS;
[6] judge on the TGS whether also have the gridding resource that need position, if having, all locate until all gridding resources that the last user of TGS is concerned about repeating step [3] ~ [5].
Test environment is to comprise that 8 256 MB of memorys, 2 * 1300MHz, operating system are the PC of Linux (Redhat 9.0), and two 1G internal memories, Pentium (R) 42.4GHz, operating systems are the server of Windows XP SP2, and these machines interconnect by the network of 100Mbps.
Register node of operation on a Windows server, grid node of operation on another Windows server.The height of the grid node resource location clue of test usefulness was 3 (comprising resource location inlet), claimed resource location inlet to be the ground floor gridding resource; The gridding resource that has direct modeling relation with resource location inlet is a second layer gridding resource; Have direct modeling relation with second layer gridding resource and with the gridding resource of ground floor gridding resource indirect association be three-layer network lattice resource.
Run user node on 8 Linux machines (can simulate a plurality of user nodes on every Linux machine).Consider that user node can be undertaken by positioning service or direct inquiry service the location of gridding resource, and positioning service is handled request by calling direct inquiry service, do not need to return intermediate object program, so locate a needed time of gridding resource always less than being positioned to the time that the target gridding resource is consumed gradually by repeatedly visiting direct inquiry service by positioning service to the user.Therefore, user node is all by the direct inquiry service gridding resource of each level on the locating node gradually in this test: at first visit the resource location clue that register node is obtained grid node, this has also just obtained the locating information (time that this stage consumed is called the positioning time of ground floor gridding resource) of resource location inlet simultaneously; By visit gridding resource location inlet, obtain the locating information of the second layer gridding resource on the grid node then; Obtain the locating information of three-layer network lattice resource again by visit second layer gridding resource.Dividing another name from the visit register node constantly, is the positioning time of second layer gridding resource and the positioning time of three-layer network lattice resource to the locating information that obtains second layer gridding resource and three-layer network lattice resource.
Can realize access rates different on the grid node by the user node number of simulating on the control Linux machine, i.e. the number of the gridding resource Location Request that receives each second of grid node.Measure grid node and handle CPU usage, user node that the gridding resource Location Request consumed and locate each level gridding resource needed average time, the result is shown in accompanying drawing 12 and accompanying drawing 13.
Can see from the data of test, grid node can be handled the gridding resource Location Request more than 120 times each second, the time that user node positions the gridding resource on the grid node is all millisecond rank (being no more than 500 milliseconds) in this trial stretch, this illustrative system has certain extensibility, has verified the validity of modeled gridding resource localization method.
Description of drawings
Fig. 1. the overall relation figure of gridding resource location.
Fig. 2. gridding resource location overview flow chart.
Fig. 3. the resource model of grid node, traversal clue and resource location clue schematic diagram.
Fig. 4. grid node is about the module and the storage distribution map of resource location.
Fig. 5. the module of register node and storage distribution map.
Fig. 6. grid node is about the workflow of resource location.
Fig. 7. the workflow of register node.
Fig. 8. the flow chart that user node positions the gridding resource on the grid node.
Fig. 9. the realization signal of cluster resource model and corresponding gridding resource thereof.
Figure 10. the different registration scenarios schematic diagrames of cluster:
● EPR (X) is the reference address of X, i.e. the locating information of X.
Figure 11. the schematic diagram that user node positions the gridding resource on the cluster:
● EPR (X) is the reference address of X, i.e. the locating information of X;
● (X Y) is the tabulation of the locating information composition of X and Y to EPRs;
● the locating information of the gridding resource of corresponding process is only returned in { process } expression;
● the locating information of all relevant gridding resources is returned in { * } expression.
Figure 12. grid node is handled the CPU usage situation that the gridding resource Location Request is consumed.
Figure 13. each level gridding resource needed average time on the user node locating node.
Embodiment
Web service resource framework WSRF (WS Resource Framework) is the up-to-date realization standard of OGSA (Open Grid ServicesArchitecture), its basic thought is " resource that state is arranged; stateless service ", comprises five technical specifications such as WS-ResourceProperties, WS-ResourceLifetime, WS-RenewableReferences, WS-ServiceGroup and WS-BaseFaults.The particular content of relevant WSRF can be with reference to http://www.oasis-open.org/committees/wsrf.
Web service distributed management WSDM (Web Services Distributed Management) comprises MUWS (ManagementUsing Web Services) and MOWS (Management Of Web Services) two parts, the former has defined and how to represent and visit manageability interface as the Web service resource, and the latter has defined and how Web service managed as resource and how to be used MUWS to describe and the visit manageability.Can particular content be with reference to http://www.oasis-open.org/committees/tc home.php? wg_abbrev=wsdm.
Apache MUSE is the project of increasing income of WSRF and WSDM, and provides workbench for the programmer further develops Web service and the grid application of following WSRF and WSDM standard.Modeled gridding resource localization method just is based on that ApacheMUSE 1.0 (http://ws.apache.org/muse) realizes.
1. the realization of shining upon between the resource model of grid node and gridding resource/Resource Properties
Grid node is modeled as the resource model of a connection.When grid node is served packing, the logic entity in the resource model is embodied as a mesh services of deferring to the WSRF standard respectively.According to the WSRF standard, a mesh services management has the gridding resource of a plurality of dynamic generations, and this has also just pointed out the mapping between gridding resource and the logic entity; Simultaneously, each modeling relation that logic entity is participated in all realizes with the Relationship of WSDM, and these Relationship will be as the Resource Properties (ResourceProperty of WSRF) of the gridding resource of logic entity correspondence.
2. example
Be realization and the location that example is set forth grid node modeling and gridding resource thereof below with the cluster.A resource model and the corresponding gridding resource of cluster are realized as shown in Figure 9.Main frame in the cluster, the process on the main frame and the network path between the cluster be counterlogic entity LE_ main frame, LE_ process and LE_ network respectively, and have " host " of one-to-many to concern between main frame and the process, so the relation of " the R_ operation has " this one-to-many is arranged between LE_ main frame and the LE_ process.For making resource model be communicated with the relation of an additional logic entity LE_ group of planes and two one-to-manies " R_ is a front end " and " R_ is connected to ".Above-mentioned 4 logic entities all correspond to the WSRF service during realization, are respectively cluster service (Singleton pattern), host services, network service and process service; 3 modeling relations are embodied as the Relationship of WSDM, are respectively " for front end ", " being connected to " and " operation has ", and they will correspondingly be added to the Resource Properties of above-mentioned WSRF service.
Grid node has a plurality of different resources location clue, thereby multiple scheme is arranged when register node is registered.Accompanying drawing 10 examples two kinds of resources location clues of cluster:
1) the Singleton resource with cluster service is a resource location inlet, can navigate to the resource of All hosts service and network service by " for front end " and " being connected to " these two Resource Properties from this resource, and, can be positioned to the resource of all process services by " operation has " this Resource Properties by the host services resource;
2) resource " path 1 " with the network service is a resource location inlet, from " 1 " of path can position the Singleton resource of cluster service by " being connected to " this Resource Properties, and then can arrive the resource of All hosts service and the resource that process is served gradually by " for front end " and " operation has " these two Resource Properties successively.
After the cluster initialization was finished and entered normal operating conditions, the set of the gridding resource of these three services of main frame, network and process was along with user's request dynamic changes.Though gridding resource is dynamically to generate, and always can locate clue by resource and come it is retrieved.Accompanying drawing 11 is a schematic flow diagram of user node location cluster online lattice resource.After user node obtained first kind of resource location clue of cluster from register node A, the result that the model element data, services ModelMetadata by cluster returns was appreciated that the contact between all gridding resources on the cluster.
1) if user node need be located the gridding resource of All hosts and process, at first extract index clue between the Singleton resource of they and cluster service (in accompanying drawing 11, being expressed as " be front end---operation has "), filtering rule as a result is set all returns (being expressed as { * } in accompanying drawing 11) for the locating information of all gridding resources, the positioning service to cluster sends request then.See shown in 4 ~ 5 in the accompanying drawing 11.
2) if user node only need be located the gridding resource of all processes, similarly, also extract index clue between the Singleton resource of they and cluster service (" be front end---operation has ") earlier, filtering rule as a result is set this moment is the locating information of only returning the gridding resource of process (in accompanying drawing 11, be expressed as { Process}), positioning service to cluster sends request again, shown in the 6-7 in the accompanying drawing 11.
3) if user node needs the gridding resource of fixer network service, extract the index clue (" being connected to ") between the Singleton resource of they and cluster service, because this index clue only comprises an element,, see 8 ~ 9 in the accompanying drawing 11 so send request to the direct inquiry service of cluster.
The hardware environment that needs: CPU1GHz or above, internal memory 256M or more than.Grid node, register node and user node may operate in the local area network (LAN) scope, also may operate in the wide area network scope of being made up of a plurality of subnets.
The software environment that grid node and register node need: the operating system, Java4.2 runtime environment, Apache Tomcat 5.0.28, the Apache Muse 1.0 that support JDK4.2
The software environment that user node needs: operating system, the Java4.2 runtime environment of supporting JDK4.2.
Claims (1)
1, modeled gridding resource localization method is characterized in that, contains following steps successively:
Step (1). in one that the supports Web service resource framework WSRF standard grid system of deferring to Open Grid Service framework OGSA, node involved in the gridding resource position fixing process is divided into grid node, user node and register node, wherein:
User node is for initiating the grid entity of gridding resource location;
Grid node provides computing capability by gridding resource to the user, and described gridding resource is meant that the Resource Owner adds the shared resource of grid system;
Register node is one or more grid nodes of keeper's appointment, be responsible for being referred to as in the maintenance network case system information of all or part of grid node, mesh services and the gridding resource of logic entity, described mesh services is deployed on the grid node, be a kind of program code of developing according to described WSRF standard, use the computing capability of shared resource to come the request of process user for mesh services with set-up function; Described gridding resource is a kind of Service Source, its attribute is exactly the task present located executing state that user's request is submitted to, this attribute is the feature that mesh services produces when the process user request, thereby has formed the mapping between logic entity and the gridding resource;
Step (2). carry out following initialization step successively by the grid node initialization module:
Step (2.1). load an XML file in the real-time storage space, the content ModelDescription of this document contains: the mapping of the mapping of the resource model of full-mesh, logic entity and gridding resource and modeling relation and gridding resource attribute, wherein:
Resource model ResourceModel={ logic entity group LogicEntities, modeling concerns Relations},
LogicEntities={ logic entity logicEntity
i, i=1,2 ..., N, N represent the logic entity number }, Relations=R (i, j)=＜logicEntity
i, logicEntity
j, j ∈ 1,2 ..., N}, i ≠ j, R (i, j) (1:N-1)=R (j, i) (N-1:1),
R (i, the j) relation that logic entity i of (1:N-1) expression can corresponding a plurality of logic entity j, R (j, i) (N-1:1) on the contrary then; Described logic entity is represented the inherent element of this grid node, and the modeling relation is represented the relation between two inherent elements, thereby has formed the mapping between modeling relation and the gridding resource attribute;
The resource model of setting up a full-mesh carries out according to the following steps:
At first, reality according to grid node, provide an initial resource model that constitutes by N logic entity and modeling relation, represent with InitRM, if this resource model is not communicated with, then this InitRM being regarded as with the logic entity is the summit, and closing with modeling is the figure on limit, find out disjoint connected subgraph in the way, m altogether;
Secondly, from each connected subgraph, select a logic entity respectively arbitrarily, in this InitRM, add a complementary logic entity and m man-to-man complementary modeling relation again, m disjoint connected subgraph is interconnected by this complementary logic entity;
Step (2.2). the maximum number of the number of initialized all kinds of gridding resources and all kinds of gridding resources in advance when grid node starts that in configuration file, is provided with according to ModelDescription and Resource Owner, instantiation mesh services and gridding resource, and preserve these information with the XML document form in the real-time storage space;
Step (2.3). make up resource location clue according to the following steps:
Step (2.3.1). gridding resource GR of picked at random is as resource location inlet from the gridding resource of instantiation;
Step (2.3.2). according to the mapping between logic entity and the gridding resource, find out the logic entity LE of gridding resource GR correspondence in resource model;
Step (2.3.3). in resource model, select one at random with the described logic entity LF of step (2.3.2) the resource model traversal clue that is root, represent with TGlue;
Step (2.3.4). selected LE of step (2.3.3) replaced with the locating information of gridding resource, and according to the mapping between modeling relation and the gridding resource attribute, non-root node among the described traversal clue TGlue is also replaced with corresponding gridding resource attribute, is the resource location clue of resource location inlet thereby obtain one with gridding resource GR;
Step (2.4). in this grid node, be provided with again with lower module:
Model element data service module ModelMetadata: the XML among the grid node visit real-time storage spatial extraction content description ModelDescription describes part, wherein:
The mapping relations of mapping relations output interface XMLObject GetModelMapping between user output is with the logic entity of XML formal description and gridding resource and between modeling relation and the gridding resource attribute;
Content description ModelDescription output interface XMLObject GetAll visit real-time storage space is to the content description ModelDescription of user's output with the statement of XML form;
Direct inquiry service module RelationService: a gridding resource locating information and gridding resource attribute with string representation that input is sent by user node with string representation, to the gridding resource locating information that user output is represented with list type, each in the tabulation all is that the XML of a gridding resource locating information describes;
Positioning service module LocationService: import the locating information of the selected gridding resource GR that sends by user node and index path LocationChain=＜RP (k of a gridding resource
1, k
2), RP (k
2, k
3) ..., RP (k
i, k
j), wherein, RP (k
1, k
j) be that corresponding modeling concerns R (k
i, k
j) the gridding resource attribute, logic entity logicEntity
K1Be the logic entity corresponding with GR; In addition, whether one group of filtering rule filterRule is used for specifying with gridding resource GR having the locating information of those gridding resources of specific modeling relation to return; This module is returned by user's appointment and gridding resource locating information that represent with list type to the user;
Step (3). on register node, be provided with lower module:
Registration service module RegisterService: a grid node sign gridSiteID that input is sent from the real-time storage spatial extraction and by grid node by grid node and a resource of this sign are located clue LG; Output is to return the information of representing with Boolean that whether succeeds in registration to grid node;
Nullify service module UnregisterService: input is sent by grid node and is identified corresponding resource location clue LG from the grid node sign of real-time storage space deletion and with this again, returns the information of the cancellation success or not of representing with Boolean to grid node;
Index service module I ndexService: the grid node sign gridSiteID that input is sent by user node, inquiry real-time storage space, the resource location hint information of this grid node from list type to user node output that represent with, each in the tabulation all are that the XML of a gridding resource location clue describes;
Step (4). carry out the modelling gridding resource localization method of initiating according to the following steps by user node:
Step (4.1). grid node is to register node registration resource behavior clue;
Step (4.2). user node is asked for the resource location clue GRLG of the own numerous target gridding node TGS that are concerned about to register node;
Step (4.3). register node is to the resource location of target gridding node described in the user node forwarding step (4.2) clue GRLG;
Step (4.4). the user asks for one of the following content of XML representation of file or multinomial to the target gridding node according to self-demand: the XML of resource model describes, the mapping of mapping between gridding resource and the logic entity and gridding resource attribute and modeling relation;
Step (4.5). user's selected grid node on the multigrid node of comforming, take out enter the mouth index path LPath between the Entry of the own gridding resource GR that is concerned about and resource location the resource location clue GRLG under the own gridding resource GR that is concerned about;
Step (4.6). whether direct modeling relation is arranged between the inlet Entry of the location of gridding resource GR and resource described in user's determining step (4.5):
If have: then this Entry and LPath as required parameter, send request to the direct inquiry service module of selected target gridding node, change step (4.7);
If do not have: filtering rule filterRule then is set, and this Entry, LPath and this filterRule send request as required parameter to the positioning service module of selected target gridding node, change step (4.7);
Step (4.7). if also have to need the gridding resource of location, all locate until this selected target gridding node TGS being gone up all gridding resources that the user is concerned about repeating step (4.3)～(4.6) then.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007101003257A CN100452726C (en) | 2007-06-08 | 2007-06-08 | Modeled network resource positioning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007101003257A CN100452726C (en) | 2007-06-08 | 2007-06-08 | Modeled network resource positioning method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101079736A true CN101079736A (en) | 2007-11-28 |
| CN100452726C CN100452726C (en) | 2009-01-14 |
Family
ID=38906982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2007101003257A Expired - Fee Related CN100452726C (en) | 2007-06-08 | 2007-06-08 | Modeled network resource positioning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100452726C (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102238188A (en) * | 2011-08-01 | 2011-11-09 | 中国人民解放军国防科学技术大学 | Application-domain-based grid resource organization method |
| CN101702020B (en) * | 2009-10-22 | 2013-03-13 | 中国农业大学 | Method and device for obtaining remotely sensed data based on grids |
| CN103098506A (en) * | 2010-07-05 | 2013-05-08 | Sk电信有限公司 | Method and device for updating a database for wireless LAN based positioning |
| CN101639531B (en) * | 2009-08-28 | 2013-08-21 | 中国农业大学 | Method for acquiring remote sensing data and device thereof |
| WO2016008086A1 (en) * | 2014-07-15 | 2016-01-21 | Microsoft Technology Licensing, Llc | Data model indexing for model queries |
| CN108596824A (en) * | 2018-03-21 | 2018-09-28 | 华中科技大学 | A kind of method and system optimizing rich metadata management based on GPU |
| US10157206B2 (en) | 2014-07-15 | 2018-12-18 | Microsoft Technology Licensing, Llc | Data retrieval across multiple models |
| US10198459B2 (en) | 2014-07-15 | 2019-02-05 | Microsoft Technology Licensing, Llc | Data model change management |
| US10423640B2 (en) | 2014-07-15 | 2019-09-24 | Microsoft Technology Licensing, Llc | Managing multiple data models over data storage system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100466657C (en) * | 2005-12-06 | 2009-03-04 | 南京邮电大学 | Access control decision-making device for grid computing environment |
| CN100440803C (en) * | 2005-12-26 | 2008-12-03 | 北京航空航天大学 | Gridding information model system and method for modeling processing gridding information |
| CN1819529A (en) * | 2006-03-16 | 2006-08-16 | 武汉理工大学 | Selecting method of grid resource model and node |
-
2007
- 2007-06-08 CN CNB2007101003257A patent/CN100452726C/en not_active Expired - Fee Related
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101639531B (en) * | 2009-08-28 | 2013-08-21 | 中国农业大学 | Method for acquiring remote sensing data and device thereof |
| CN101702020B (en) * | 2009-10-22 | 2013-03-13 | 中国农业大学 | Method and device for obtaining remotely sensed data based on grids |
| CN103098506B (en) * | 2010-07-05 | 2016-02-24 | Sk电信有限公司 | Upgrade the method and apparatus be used for based on the database of the location of WLAN |
| CN103098506A (en) * | 2010-07-05 | 2013-05-08 | Sk电信有限公司 | Method and device for updating a database for wireless LAN based positioning |
| CN102238188B (en) * | 2011-08-01 | 2014-05-07 | 中国人民解放军国防科学技术大学 | Application-domain-based grid resource organization method |
| CN102238188A (en) * | 2011-08-01 | 2011-11-09 | 中国人民解放军国防科学技术大学 | Application-domain-based grid resource organization method |
| WO2016008086A1 (en) * | 2014-07-15 | 2016-01-21 | Microsoft Technology Licensing, Llc | Data model indexing for model queries |
| CN105518670A (en) * | 2014-07-15 | 2016-04-20 | 微软技术许可有限责任公司 | Data model indexing for model queries |
| US10140323B2 (en) | 2014-07-15 | 2018-11-27 | Microsoft Technology Licensing, Llc | Data model indexing for model queries |
| US10157206B2 (en) | 2014-07-15 | 2018-12-18 | Microsoft Technology Licensing, Llc | Data retrieval across multiple models |
| US10198459B2 (en) | 2014-07-15 | 2019-02-05 | Microsoft Technology Licensing, Llc | Data model change management |
| US10423640B2 (en) | 2014-07-15 | 2019-09-24 | Microsoft Technology Licensing, Llc | Managing multiple data models over data storage system |
| CN105518670B (en) * | 2014-07-15 | 2021-09-07 | 微软技术许可有限责任公司 | Data model indexing for model queries |
| CN108596824A (en) * | 2018-03-21 | 2018-09-28 | 华中科技大学 | A kind of method and system optimizing rich metadata management based on GPU |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100452726C (en) | 2009-01-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101079736A (en) | Modeled network resource positioning method | |
| Karczewski et al. | Systematic single-variant and gene-based association testing of thousands of phenotypes in 394,841 UK Biobank exomes | |
| US9778967B2 (en) | Sophisticated run-time system for graph processing | |
| TWI333621B (en) | Facilitating allocation of resources in a heterogeneous computing environment | |
| US9304835B1 (en) | Optimized system for analytics (graphs and sparse matrices) operations | |
| CN1252592C (en) | Communications services provisioning method and apparatus and object programming language for developing provisioning models | |
| Cai et al. | A transparent approach of enabling SaaS multi-tenancy in the cloud | |
| CN1879082A (en) | Method and apparatus for composing software | |
| CN1759397A (en) | Structured indexes on results of function applications over data | |
| CN1975720A (en) | Data tapping system based on Wcb and control method thereof | |
| CN1804840A (en) | Data access layer class generator | |
| CN1914630A (en) | Behavior based multi-agent systems as data types. | |
| CN1647042A (en) | Methods for customizing software abstractions | |
| CN1755617A (en) | Entity domains | |
| CN1890643A (en) | Complex computation across heterogenous computer systems | |
| CN1208482A (en) | Distributed processing | |
| CN1961288A (en) | Methods and apparatus for accessing process control data | |
| CN101064641A (en) | Method and system for attribute based index and/or query of Web service resource | |
| CN1716249A (en) | System and method for delayed fetching of designated members of a user defined type | |
| CN104050042A (en) | Resource allocation method and resource allocation device for ETL (Extraction-Transformation-Loading) jobs | |
| CN1866283A (en) | System and method for implementing regular system triggering | |
| CN1866250A (en) | Method and system for managing system data | |
| CN1823322A (en) | Shared code caching method and apparatus for program code conversion | |
| CN1662011A (en) | System and method for building component applications using metadata defined mapping between message and data domains | |
| CN1258704C (en) | Method for using synergic multiple computer engines based on gridding environment |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090114 Termination date: 20160608 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |