CN106254471A - Resource United Dispatching method and system under a kind of isomery cloud environment - Google Patents
Resource United Dispatching method and system under a kind of isomery cloud environment Download PDFInfo
- Publication number
- CN106254471A CN106254471A CN201610644721.5A CN201610644721A CN106254471A CN 106254471 A CN106254471 A CN 106254471A CN 201610644721 A CN201610644721 A CN 201610644721A CN 106254471 A CN106254471 A CN 106254471A
- Authority
- CN
- China
- Prior art keywords
- node
- request
- service
- calculating
- operating system
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/56—Provisioning of proxy services
- H04L67/566—Grouping or aggregating service requests, e.g. for unified processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of resource United Dispatching framework and method under isomery cloud environment, this framework includes: scheduling controller, at least one calculating node and the calculating each self-corresponding request agent node of node；The first service request that scheduling controller sends for receiving terminal apparatus, and send second service request according to first service request to request agent node, it is that terminal unit distribution calculates node that first service request is used for asking scheduling controller, and second service request provides, for terminal unit, the service of calculating for indicating the calculating node of different operating system；Request agent node for being converted to the service command in different operating system and by calculating node corresponding for service command transmission to different operating system by second service request, calculating node and be used for according to service command being that terminal unit provides the service of calculating, operating system includes cloud operating system and non-cloud operating system.The problem being managed collectively scheduling that existing network does not support the hardware of cloud can be solved.
Description
Technical field
The present invention relates to field of cloud calculation, particularly relate to resource United Dispatching method and system under a kind of isomery cloud environment.
Background technology
Along with the development of computer software and hardware, its application is also deep into the various calculating of all trades and professions, designs and manage
In reason, the work efficiency of each enterprise and mechanism is made to be greatly improved.In recent years, occur in that especially a kind of cloud computing (English:
Cloud computing), (English: concept cloud platforms), cloud computing is based on the Internet relevant to cloud platform
Increase, use and the delivery mode of service, is usually directed to be provided by the Internet the most easily extend and the most virtualized
Resource.Cloud is a kind of metaphor saying of network, the Internet.Past the most often represents telecommunications network with cloud, is the most also used for
Represent the abstract of the Internet and underlying infrastructure.Narrow sense cloud computing refers to payment and the use pattern of IT infrastructure, refers to pass through
Network obtains resource requirement with on-demand, easy extension way；Broad sense cloud computing refers to payment and the use pattern of service, refers to by net
Network obtains required service with on-demand, easy extension way.It is relevant with software, the Internet that this service can be IT, it is possible to is it
He services.It means that computing capability also can be circulated by the Internet as a kind of commodity.
Cloud computing is try to utilize cloud computing to carry due to its manageability, advantage, numerous enterprises or the operator such as scalable
Rise competitiveness.But, enterprise or operator when disposing solution, the problem being faced with infrastructure isomery.I.e. enterprise or
There are some old hardware in operator's existing network, some offer physical machine (English: Physical Machine, PM) of these hardware, has
Virtual machine (English: Virtual Machine, VM) is provided, is not to support cloud computing.If enterprise or operator adopt
Dispose by full non-cloudization, then cannot have the advantage of cloud computing resources scheduling and management.If using full cloudization to dispose, abandon existing
Net a lot of old hardware and use new hardware cost too high.
Accordingly, it would be desirable to a kind of framework realize existing can cloud and can not the multiplexing of VM, PM resource of cloud, and be similar to
United Dispatching in cloud computing manages.
Summary of the invention
Embodiments provide resource United Dispatching method and system under a kind of isomery cloud environment, it is possible to resolve existing network is not
The problem supporting the unified management scheduling of the hardware of cloud.
First aspect, it is provided that resource United Dispatching framework under a kind of isomery cloud environment, including: scheduling controller, at least
One calculates node and each self-corresponding request agent node of described calculating node, described scheduling controller and described request agency
Node connects, and described request agent node is connected with described calculating node；Described scheduling controller is sent out for receiving terminal apparatus
The first service request sent, and send second service request, institute according to the request of described first service to described request agent node
Stating first service request for asking described scheduling controller is that the distribution of described terminal unit calculates node, and described second service please
Ask for indicating the calculating node of different operating system to provide, for described terminal unit, the service of calculating；Described request agent node is used
In the request of described second service being converted to the service command in different operating system and described service command being sent to different
The calculating node that operating system is corresponding, described calculating node is for providing calculating according to described service command for described terminal unit
Service, described operating system includes cloud operating system and non-cloud operating system.
By applying above-mentioned framework, it is possible to resolve existing network does not support the problem that the unified resource of the hardware of cloud is dispatched.
In conjunction with first aspect, in the first implementation of first aspect, described framework also include manager, at least one
Individual control node and each self-corresponding control agent node of described control node, described manager is for receiving what manager sent
First management request, described first management request is used for asking described manager to be managed described calculating node；Described pipe
Reason device is additionally operable to send the second management request, described second management according to described first management request to described control agent node
Request is used for indicating described control node to be managed described calculating node；Described control agent node is for by described second
Management request is converted to the administration order in different operating system and sends described administration order to different operating system corresponding
Control node, described control node is for being managed described calculating node according to described administration order.
By applying above-mentioned framework, it is possible to resolve the problem that existing network does not support the unified calculation resource management of the hardware of cloud.
In conjunction with the first implementation of first aspect, in the second implementation of first aspect, described control is saved
Point is additionally operable to after having performed described administration order send the first execution result, described control agent to described control agent node
Node is additionally operable to perform result by described first and reports described manager, and described manager is additionally operable to perform according to described first
Result updates and controls node state table.
By applying above-mentioned framework, manager can collect each information controlling node, controls node according to each
Information is managed the rational management of request.
In conjunction with the first implementation of first aspect, or the second implementation of first aspect, in first aspect
In the third implementation, described second management request is carried the mark of target control node, described target control node
Described second management request is converted to meet for described control agent node and grasps corresponding to described target control node by mark
Make the administration order of the command format of system, and the management of the command format by operating system corresponding to described target control node
Order sends to described target control node.
In conjunction with first aspect, or any one implementation of first aspect, in the 4th kind of implementation of first aspect
In, described calculating node is additionally operable to send second to described request agent node after having performed described service command and performs knot
Really, described request agent node is additionally operable to perform result by described second and reports described scheduling controller, described scheduling controlling
Device is additionally operable to perform result according to described second and updates calculating node state table.
By applying above-mentioned framework, scheduling controller can collect each information calculating node, calculates joint according to each
The information of point carries out the rational management of service request.
In conjunction with first aspect, or any one implementation of first aspect, in the 5th kind of implementation of first aspect
In, the request of described second service is carried the mark of target computing nodes, the mark of described target computing nodes is asked for described
Agent node is asked to be converted to meet the order lattice of operating system corresponding to described target computing nodes by the request of described second service
The service command of formula, and the service command of the command format of operating system corresponding to described target computing nodes is sent to described
Target computing nodes.
Second aspect, it is provided that a kind of resource United Dispatching method under isomery cloud environment, the method includes:
The first service request that scheduling controller receiving terminal apparatus sends, the request of described first service is used for asking described
Scheduling controller is that the distribution of described terminal unit calculates node；
Described scheduling controller sends second service request according to the request of described first service to destination request agent node,
The request of described second service is used for indicating target computing nodes to provide, for described terminal unit, the service of calculating；
It is right that the request of described second service is converted to meet described target computing nodes institute by described destination request agent node
Answer the service command of the command format of operating system；
Described service command is sent to described target computing nodes by described destination request agent node；
Described target computing nodes provides, for described terminal unit, the service of calculating according to described service command.
By performing above-mentioned steps, it is possible to resolve existing network does not support the problem that the unified resource of the hardware of cloud is dispatched.
In conjunction with second aspect, in the first implementation of second aspect, described scheduling controller receiving terminal apparatus
Before the first service request sent, also include:
Manager receives the first management request, carries the mark of described target computing nodes in described first management request,
Described first management request is used for asking described manager to start described target computing nodes；
Described manager sends the second management request according to described first management request to target control agent node, described
Second management request is used for indicating target control node to start described target computing nodes；
It is right that described second management request is converted to meet described target control node institute by described target control agent node
Answer the administration order of the command format of operating system；
Described administration order is sent to described target control node by described target control agent node；
Described target control node starts described target computing nodes according to described administration order.
By performing above-mentioned steps, it is possible to resolve the problem that existing network does not support the unified calculation resource management of the hardware of cloud.
In conjunction with the first implementation of second aspect, in the second implementation of second aspect, described target control
After node processed starts described target computing nodes according to described administration order, also include:
Described target control node sends the first execution result to described target control agent node；
Described target control agent node performs result by described first and is sent to described manager；
Described manager performs result according to described first and updates control node state table, and described control node state table is used
From at least one control agent node, described target control agent node is determined in described manager.
By performing above-mentioned steps, manager can collect each information controlling node, controls node according to each
Information is managed the rational management of request.
In conjunction with the first implementation of second aspect, or the second implementation of second aspect, in second aspect
In the third implementation, carrying the mark of described target control node in described second management request, described target control saves
The mark of point determines described target control node for described target control agent node.
In conjunction with second aspect, or any one implementation of second aspect, in the 4th kind of implementation of second aspect
In, after described target computing nodes provides, for described terminal unit, the service of calculating according to described service command, also include:
Described target computing nodes sends the second execution result to described request agent node；
Described request agent node performs result by described second and is sent to described scheduling controller；
Described scheduling controller performs result according to described second and updates calculating node state table, described calculating node state
Table determines described target computing nodes for described scheduling controller from least one calculating node.
By performing above-mentioned steps, scheduling controller can collect each information calculating node, calculates joint according to each
The information of point carries out the rational management of service request.
In conjunction with second aspect, or any one implementation of second aspect, in the 5th kind of implementation of second aspect
In, the request of described second service is carried the mark of described target computing nodes, described target computing nodes is used for described target
Request agent node determines described target computing nodes.
Accompanying drawing explanation
For the technical scheme being illustrated more clearly that in the embodiment of the present invention, in embodiment being described below required for make
Accompanying drawing be briefly described.
Fig. 1 is resource United Dispatching configuration diagram under the isomery cloud environment that the embodiment of the present invention one provides；
Fig. 2 is resource United Dispatching configuration diagram under the isomery cloud environment that the embodiment of the present invention two provides；
Fig. 3 is resource United Dispatching configuration diagram under the isomery cloud environment that the embodiment of the present invention three provides；
Fig. 4 is resource United Dispatching configuration diagram under the isomery cloud environment that the embodiment of the present invention four provides；
Fig. 5 is the schematic flow sheet that under a kind of isomery cloud environment that the embodiment of the present invention provides, resource unifies tune method；
Fig. 6 is the schematic flow sheet calculating node start-up course that the embodiment of the present invention provides.
Detailed description of the invention
The term that embodiments of the present invention part uses is only used for explaining the specific embodiment of the present invention, rather than
It is intended to limit the present invention.Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out
It is explicitly described.
Refer to Fig. 1, be resource United Dispatching configuration diagram under the isomery cloud environment of the embodiment of the present invention one offer, should
Resource United Dispatching framework includes scheduling controller 101, calculates node 102, calculating node 103, calculating node 104, calculating joint
Point 105 and request agent node (English: Unified Request Process Node, URPN) 106, the embodiment of the present invention with
Illustrate as a example by the corresponding same request agent node 106 of multiple calculating nodes.Wherein, scheduling controller 101 and request agency
Node 106 is connected, and request agent node 106 calculates node with above-mentioned 4 respectively and is connected.Calculate the behaviour of node 102 correspondence
Being cloud operating system as system, described cloud operating system can include but not limited to: OpenStack operating system,
CloudStack operating system, Elaster cloud operating system, in embodiments of the present invention, corresponding to calculate node 102
Illustrate as a example by OpenStack operating system, and calculating node 102 is virtual machine (English: Virtual Machine, VM)；
The operating system calculating node 103 correspondence is non-cloud operating system, and described non-cloud operating system includes but not limited to:
Simulator operating system, Simulator is a kind of virtual machine management system from the non-cloud ground, in the embodiment of the present invention
In, illustrate as a example by calculating the corresponding Simulator operating system of node 103, and calculating node 103 is virtual machine；Calculate
The operating system of node 104 correspondence is cloud operating system, in embodiments of the present invention, corresponding to calculate node 104
Illustrate as a example by OpenStack operating system, and calculate node 104 be physical machine (English: Physical Machine,
PM)；The operating system calculating node 105 correspondence is non-cloud operating system, in embodiments of the present invention, to calculate node 105
Illustrate as a example by corresponding Simulator operating system, and calculating node 105 is physical machine.Wherein, virtual machine or physical machine
In different application program (English: Application, APP) can be installed.
Request agent node 106 is all cloud operating systems, the non-cloud operating systems such as OpenStack, Simulator
Calculating the agent node of node, the function of request agent node 106 indispensability includes: receive the unification that scheduling controller 101 issues
End-user services request, scheduling controller 101 is issued service request be converted in different operating system service life
Make and issue, collect each information calculating node (such as procotol (English: Internet Protocol, IP) address,
Performance etc.), and each is calculated the information reporting of node to scheduling controller 101.Can be realized by request agent node 106
No matter the same scheduling problem of different operating system, therefore, be cloud operating system, or non-cloud operating system, all can portion
It is deployed in resource United Dispatching framework it is achieved thereby that the multiplexing of non-cloud VM/PM equipment, decreases equipment cost.
Concrete, the process that the request of servicing terminal unit transmission is scheduling is: scheduling controller 101 receives terminal
The first service request that equipment 200 sends, it is terminal unit 200 points that the request of this first service is used for asking scheduling controller 101
Joining calculating node, wherein, terminal unit 200 can send this first service by wired mode or wireless mode and ask.Scheduling
Controller 101 generates second service request according to the request of described first service, and the request of this first service is asked with this second service
Form and content can be the same or different, scheduling controller 101, to destination request agent node, i.e. asks agent node
106 send the request of these second services, and the request of this second service includes the mark of the target computing nodes that scheduling controller 101 determines
Knowing, the request of this second service is used for indicating target computing nodes to provide, for terminal unit 200, the service of calculating.Request agent node
After 106 receive the request of described second service, from described second service is asked, obtain the mark of described target computing nodes,
The request of described second service is converted to meet the service life of the command format of operating system corresponding to described target computing nodes
Order, concrete, request agent node 106 can the mapping table of mark of mark and operating system of pre-recorded calculating node,
After request agent node 106 gets the mark of described target computing nodes, look into according to the mark of described target computing nodes
Ask described mapping table, i.e. can determine which the operating system that described target computing nodes is corresponding is, and then by described second clothes
Business request is converted to the command format of operating system corresponding to described target computing nodes.Such as, if described target computing nodes
For calculate node 102, the most described target computing nodes be designated 102, operating system corresponding to described target computing nodes is
OpenStack operating system, the request of described second service is converted to OpenStack operating system pair by request agent node 106
The service command answered.
Request agent node 106 determines described target meter according to the mark (such as identifying 102) of described target computing nodes
Operator node (such as calculates node 102), and the described service command after conversion is sent to described target computing nodes.
After described target computing nodes (such as calculating node 102) receives described service command, according to described service life
Make and provide, for terminal unit 200, the service of calculating.
Optionally, described target computing nodes can also ask agent node by being back to when servicing the fructufy of user
106, the result of return includes but not limited to: the most service, and services which terminal unit, services which user, current mesh
Mark calculates the information such as the state of node.Request agent node 106 is additionally operable to the information conversion returned by described target computing nodes
Becoming the form mated with scheduling controller 101, and then be back to scheduling controller 101, scheduling controller 101 is according to the knot returned
Fruit updates the calculating node state table of self storage.In this calculating node state table, record has each to calculate the information of node, bag
Include but be not limited to: the most service, service which terminal unit, service which user, the state of current goal calculating node.
Scheduling controller 101 according to record calculatings node state table i.e. would know that each calculating node state, therefore,
After the first service request that scheduling controller 101 receiving terminal apparatus 200 sends, scheduling controller 101 can be inquired about described
Calculate node state table, combine dispatching algorithm according to each status information calculating node and determine which is calculated node makees
For described target computing nodes.Concrete, the dispatching algorithm of scheduling controller 101 can be the load balancing of each calculating node,
Can also be (English: User Profile), according to the feature of different terminals user in conjunction with the user profile of big data component
And the service request that terminal use is sent by demand is dispatched to suitably calculate node.Such as, if terminal use 1 to service matter
Amount requires higher, and scheduling controller 101 is asked to be assigned to PM (such as calculating node 104), if terminal use 2 is to clothes
Business prescription is the highest, and scheduling controller 101 is asked to be assigned to VM (such as calculating node 102).
Although it should be noted that in embodiments of the present invention, being said as a example by the quantity calculating node is 4
Bright, but, the most only it should be apparent to a person skilled in the art that the quantity calculating node included by resource United Dispatching framework
It is limited to 4, it is also possible to be other quantity, such as 1,10,20, etc..
Refer to Fig. 2, be resource United Dispatching configuration diagram under the isomery cloud environment of the embodiment of the present invention two offer, should
Resource United Dispatching framework is except including scheduling controller 101, calculating node 102, calculating node 103, calculating node 104, meter
Beyond operator node 105, request agent node 106, also include: manager 107, control node (English: Control Node, CN)
108, control node 109, control node 110, control node 111 and control agent node (English: Unified Control
Node, UCN) 112, wherein, scheduling controller 101, calculating node 102, calculating node 103, calculating node 104, calculating node
105 and the 26S Proteasome Structure and Function of request agent node 106 can be found in the associated description of embodiment illustrated in fig. 1, here is omitted.
Manager 107 is connected with control agent node 112, and control agent node 112 controls node with each and is connected.The present invention is real
Execute example to illustrate as a example by the corresponding same control agent node 112 of multiple control nodes, wherein, for every kind of operating system,
There is the corresponding node that controls to manage each calculating node, cloud operating system as corresponding in above-mentioned control node 108:
OpenStack operating system, the corresponding non-cloud operating system of control node 109: Simulator operating system, control node 110
Corresponding cloud operating system: OpenStack operating system, controls node 111 corresponding non-cloud operating system: Simulator behaviour
Make system.Furthermore, it is necessary to explanation, the node that controls of every kind of operating system can be to the multiple of operating system calculating joint
Point, the embodiment of the present invention illustrates as a example by each control node each a corresponding calculating node.Node computed as above
The 102 corresponding nodes 108 that control, calculating node 103 is corresponding controls node 109, calculates node 104 corresponding control node 110, calculates
Node 105 is corresponding controls node 111.
Although it should be noted that not shown, each controls node and each calculating node is provided with interface, so that control
Node processed is controlled calculating node.
Each function controlling node is each calculating node (VM or PM) of management, and this management includes but not limited to: create
VM, deletion VM, startup PM, stopping PM.Control agent node 112 is all clouds such as OpenStack, Simulator, non-cloud
Operating system controls the agent node of node, and the function of control agent node 112 indispensability includes: receive what manager 107 issued
Unified management request, is converted into the corresponding administration order controlling operating system corresponding to node by unified management request and is issued to
Control node, collect each information controlling node, and each control node is reported to manager 107.
Concrete, to each process calculating node overall calculation resource management it is: operation maintenance personnel (i.e. manager) send
First management is asked, and carries the mark of a certain target computing nodes in this first management request, and this first management request is used for please
Manager 107 is asked to start described target computing nodes.Manager 107 receives this first management request, please according to this first management
Seeking survival into the second management request, this second management request is used for indicating target control node to start described target computing nodes.Should
First management request can be the same or different with form and the content of this second management request.Manager 107 is to target control
Agent node processed, i.e. control agent node 112 send the second management request, and described second management request includes manager 107
The mark of the target control node determined and the mark of target computing nodes, control agent node 112 receives described second pipe
After reason request, from described second management request, obtain the mark of described target control node, by described second management request
Be converted to meet the administration order of the command format of operating system corresponding to described target control node, concrete, control agent
Node 112 can the mapping table of mark of mark and operating system of pre-recorded control node, when control agent node 112 obtains
After getting the mark of described target control node, inquire about described mapping table according to the mark of described target control node, can be really
Make which operating system corresponding to described target control node is, and then described second management request is converted to described target
Control the command format of operating system corresponding to node.Such as, if described target control node is for controlling node 108, then described
Target control node be designated 108, operating system corresponding to described target control node is OpenStack operating system, control
Described second management request is converted to the administration order that OpenStack operating system is corresponding by agent node 112 processed.
Optionally, described first management request can also be carried the mark of a certain target control node that manager selects
Know.
Control agent node 112 determines described target control according to the mark (such as identifying 108) of described target control node
Node processed (such as controls node 108), and the described administration order after conversion is sent to described target control node.
After described target control node (such as controlling node 108) receives described administration order, carry out calculating node
Management, such as, if described administration order is for creating VM order, then controls node 108 and creates a new VM, if described management life
Order for starting PM order, then controls node 108 and starts a PM, further, can also carry the mark of PM in this administration order
Know, control node 108 and start corresponding PM according to the mark of PM.
Optionally, described target control node will can also perform to return during fructufy after having performed described administration order
To control agent node 112, the result of return includes whether VM creates successfully, and whether PM starts or stops successfully etc..Control generation
The result that target control node returns can also be converted into suitable form by reason node 112, is back to manager 107, management
Device 107 updates the control node state table of self storage according to the result returned.In this control node state table, record has each
Control the information of node, include but not limited to: currently have how many VM/PM running on each control node, also can create how many
VM/PM etc..
According to the control node state table of record, manager 107 i.e. would know that each controls the state of node, therefore, at pipe
After reason device 107 receives the first management request that manager sends, described control node state table can be inquired about, control according to each
The status information of node processed combines dispatching algorithm and determines which control node as described target control node.Specifically
, the dispatching algorithm of manager 107 can be the load balancing of each control node, it is also possible to be the meter that can create of each control node
The quantity of operator node.Such as, control node 108 and also can create 3 calculating nodes, control node 109 and also can create 0 calculating joint
Point, controls node 110 and also can create 1 calculating node, controls node 111 and also can create 2 calculating node, then manager 107
Control node 108 is defined as described target control node, thus controls node 108 and create a new calculating node.
Although it should be noted that in embodiments of the present invention, being said as a example by the quantity controlling node is 4
Bright, but, the most only it should be apparent to a person skilled in the art that the quantity controlling node included by resource United Dispatching framework
It is limited to 4, it is also possible to be other quantity, such as 1,10,20, etc..
Refer to Fig. 3, be resource United Dispatching configuration diagram under the isomery cloud environment of the embodiment of the present invention three offer, should
Resource United Dispatching framework includes scheduling controller 301, calculates node 302, calculating node 303, calculating node 304, calculating joint
Point 305, request agent node 306, request agent node 307, request agent node 308, request agent node 309, with enforcement
Unlike example one, the embodiment of the present invention illustrates as a example by the corresponding different request agent node of each calculating node, tool
Body, calculate node 302 corresponding requests agent node 306, calculate node 303 corresponding requests agent node 307, calculate node
304 corresponding requests agent nodes 308, calculate node 305 corresponding requests agent node 309, and each request agent node is on upper strata
Scheduling controller 101 is reciprocity.Wherein, scheduling controller 101 is connected with each request agent node, and each asks generation
The calculating node managing node the most corresponding is connected.In the embodiment of the present invention, calculate node, the function of request agent node
Identical with embodiment one, here is omitted.
Concrete, the process that the request of servicing terminal unit transmission is scheduling is: scheduling controller 301 receives terminal
The first service request that equipment 200 sends, it is described terminal unit that the request of described first service is used for asking scheduling controller 301
Distribution calculates node, and scheduling controller 301 combines the calculating node state table of self storage, determines from multiple calculating nodes
Target computing nodes (is assumed to be calculating node 302), and generates second service request, scheduling controller according to first service request
301 determine the destination request agent node (i.e. request agent node 306) that this target computing nodes is corresponding, concrete, scheduling control
In device 301 processed, storage has each to calculate the mapping table of node and each request agent node, i.e. would know that mesh by this mapping table
What mark calculating node was corresponding is which asks agent node.Scheduling controller 301 is to destination request corresponding to target computing nodes
Agent node sends second service request, and it is that terminal unit 200 carries that the request of described second service is used for indicating target computing nodes
For calculating service, after described destination request agent node receives the request of described second service, described second service is asked
Be converted to meet the service command of the command format of operating system corresponding to described target computing nodes, and then by the institute after conversion
State service command and be sent to described target computing nodes, the mark of carried terminal equipment 200 in described service command.
After described target computing nodes (such as calculating node 302) receives described service command, according to described service life
Order services for terminal unit 200.
Optionally, described target computing nodes can also will be back to described destination request generation when servicing the fructufy of user
Reason node, the result of return includes but not limited to: the most service, and services which terminal unit, services which user, current
The information such as the state of target computing nodes.Described destination request agent node is additionally operable to the letter returned by described target computing nodes
Breath is converted into the form mated with scheduling controller 301, and then is back to scheduling controller 301, and scheduling controller 301 is according to returning
The result returned updates the calculating node state table of self storage.In this calculating node state table, record has each to calculate the letter of node
Breath, includes but not limited to: the most service, service which terminal unit, service which user, current goal calculates node
State.
Scheduling controller 301 according to record calculatings node state table i.e. would know that each calculating node state, therefore,
After the first service request that scheduling controller 301 receiving terminal apparatus 200 sends, scheduling controller 301 can be inquired about described
Calculate node state table, combine dispatching algorithm according to each status information calculating node and determine which is calculated node makees
For described target computing nodes.Concrete, the dispatching algorithm of scheduling controller 301 can be the load balancing of each calculating node,
Can also be in conjunction with the user profile of big data component, terminal use is sent by feature and demand according to different terminals user
Service request be dispatched to suitably calculate node.
Although it should be noted that in embodiments of the present invention, being said as a example by the quantity calculating node is 4
Bright, but, the most only it should be apparent to a person skilled in the art that the quantity calculating node included by resource United Dispatching framework
It is limited to 4, it is also possible to be other quantity, such as 1,10,20, etc..
Refer to Fig. 4, be resource United Dispatching configuration diagram under the isomery cloud environment of the embodiment of the present invention four offer, should
Resource United Dispatching framework is except including scheduling controller 301, calculating node 302, calculating node 303, calculating node 304, meter
Operator node 305, request agent node 306, request agent node 307, request agent node 308, request agent node 309 with
Outward, also include: manager 310, control node 311, control node 312, control node 313, control node 314, control agent
Node 315, control agent node 316, control agent node 317 and control agent node 318, unlike embodiment two,
The embodiment of the present invention illustrates as a example by the corresponding different control agent node of each control node, concrete, controls node
311 corresponding control agent nodes 315, control the corresponding control agent node 316 of node 312, control the corresponding control agent of node 313
Node 317, controls the corresponding control agent node 318 of node 314, and each control agent node is in the manager 310 on upper strata
Equity.Wherein, manager 310 is connected with each control agent node, and each control agent node is the most corresponding
Control node to be connected.Scheduling controller 301, calculate node 302, calculate node 303, calculate node 304, calculate node 305,
Request agent node 306, request agent node 307, request agent node 308 and the structure of request agent node 309 and merit
Can be found in the associated description of embodiment illustrated in fig. 3, here is omitted.
Concrete, to each process calculating node overall calculation resource management it is: operation maintenance personnel (i.e. manager) send
First management request, manager 310 receives this first management request, in conjunction with the control node state table of self storage, from multiple
Control node is determined target control node (being assumed to be control node 311), and according to described first management request generation the
Two management requests, manager 310 determines target control agent node (the i.e. control agent node that this target control node is corresponding
315), concrete, in manager 310, storage has each to control the mapping table of node and each control agent node, is reflected by this
Firing table i.e. would know that target control node corresponding be which control agent node.Manager 310 is to target control agent node
Send the second management request, after described target control agent node receives described second management request, by described second pipe
Reason request is converted to meet the administration order of the command format of operating system corresponding to described target control node, and then will conversion
After described administration order be sent to described target control node.
After described target control node (such as controlling node 311) receives described administration order, carry out calculating node
Management, such as, if described administration order is for creating VM order, then controls node 311 and creates a new VM, if described management life
Order for starting PM order, then controls node 311 and starts a PM, further, can also carry the mark of PM in this administration order
Know, control node 311 and start corresponding PM according to the mark of PM.
Optionally, described target control node will can also perform to return during fructufy after having performed described administration order
To described target control agent node, the result of return includes whether VM creates successfully, and whether PM starts or stops successfully etc..Institute
State target control agent node and the result that target control node returns can also be converted into suitable form, be back to manager
310, manager 310 updates the control node state table of self storage according to the result returned.This control node state table is remembered
Record has each to control the information of node, includes but not limited to: currently has how many VM/PM running on each control node, also may be used
Create how many VM/PM etc..
According to the control node state table of record, manager 310 i.e. would know that each controls the state of node, therefore, at pipe
After reason device 310 receives the first management request that manager sends, described control node state table can be inquired about, control according to each
The status information of node processed combines dispatching algorithm and determines which control node as described target control node.Specifically
, the dispatching algorithm of manager 310 can be the load balancing of each control node, it is also possible to be the meter that can create of each control node
The quantity of operator node.
Although it should be noted that in embodiments of the present invention, being said as a example by the quantity controlling node is 4
Bright, but, the most only it should be apparent to a person skilled in the art that the quantity controlling node included by resource United Dispatching framework
It is limited to 4, it is also possible to be other quantity, such as 1,10,20, etc..
It should be noted that resource United Dispatching framework shown in above-mentioned Fig. 1 to Fig. 4 is only used as several citing, specifically should
In with, it is also possible to combine concrete condition adjustresources United Dispatching framework, such as, in resource United Dispatching framework, all of
Calculate the corresponding common request agent node of node, and for calculating the control node corresponding to node, each control
The most corresponding single control agent node of node.
Infrastructure (such as VM or PM) in existing network is not relied on by resource United Dispatching framework shown in above-mentioned Fig. 1 to Fig. 4,
No matter it is the infrastructure in cloud operating system, or the infrastructure of non-cloud operating system, all can be deployed to resource system
In one scheduling architecture, it is achieved to fully reusing of non-cloud infrastructure, reduce equipment cost, remarkable in economical benefits.Solve
Existing network does not support the unified management scheduling problem of the hardware of cloud.Further, resource United Dispatching framework operates with existing cloudization
System, virtualization system and hardware management system compatible.The user that performance requirement is high can be directly deployed on PM, it is not necessary to worry
The problem of management thus brought or the performance loss being deployed in VM.Additionally, scheduling controller grasps user profile, configurable many
Plant dispatching algorithm.
Refer to Fig. 5, be the flow process of resource United Dispatching method under a kind of isomery cloud environment of embodiment of the present invention offer
Schematic diagram, the method includes but not limited to following steps.
The first service request that S501, scheduling controller receiving terminal apparatus send, the request of described first service is for asking
Seeking described scheduling controller is that the distribution of described terminal unit calculates node.
S502, described scheduling controller send second service according to the request of described first service to destination request agent node
Request, the request of described second service is used for indicating target computing nodes to provide, for described terminal unit, the service of calculating.
The request of described second service is converted to meet described target computing nodes by S503, described destination request agent node
The service command of the command format of corresponding operating system.
Described service command is sent to described target computing nodes by S504, described destination request agent node.
S505, described target computing nodes provide, for described terminal unit, the service of calculating according to described service command.
Optionally, the request of described second service is carried the mark of described target computing nodes, described target computing nodes
Described target computing nodes is determined for described destination request agent node.
Optionally, refer to Fig. 6, be the schematic flow sheet starting described calculating node.Described scheduling controller receives eventually
Before the first service request that end equipment sends, further comprising the steps of S506 to S510:
S506, manager receive the first management request, and generate the second management request, institute according to described first management request
Stating the mark carrying described target computing nodes in the first management request, described first management request is used for asking described manager
Start described target computing nodes.
S507, described manager send the second management according to described first management request to target control agent node please
Asking, described second management request is used for indicating target control node to start described target computing nodes.
Described second management request is converted to meet described target control node by S508, described target control agent node
The administration order of the command format of corresponding operating system.
Described administration order is sent to described target control node by S509, described target control agent node.
S510, described target control node start described target computing nodes according to described administration order.
Optionally, described second management request is carried the mark of described target control node, described target control node
Mark determine described target control node for described target control agent node.
Optionally, after described target control node starts described calculating node according to described administration order, also include: institute
State target control node and send the first execution result to described target control agent node；Described target control agent node is by institute
State the first execution result and be sent to described manager；Described manager performs result according to described first and updates control node state
Table, described control node state table determines described target control for described manager from least one control agent node described
Agent node processed.
Optionally, described target computing nodes provides, for described terminal unit, the service of calculating according to described service command
After, also include: described target computing nodes sends the second execution result to described request agent node；Described request agent node
Perform result by described second and be sent to described scheduling controller；Described scheduling controller performs result according to described second and updates
Calculating node state table, it is true from least one calculating node described that described calculating node state table is used for described scheduling controller
Fixed described target computing nodes.
It should be noted that the function of each functional module can in resource United Dispatching method described by the embodiment of the present invention
According to the associated description in above-mentioned Fig. 1 or Fig. 2 or Fig. 3 or embodiment illustrated in fig. 4, here is omitted.
In sum, the infrastructure (such as VM or PM) in existing network is not relied on by above-mentioned resource United Dispatching framework, no
Pipe is the infrastructure in cloud operating system, or the infrastructure of non-cloud operating system, all can be deployed to resource unified
In scheduling architecture, it is achieved to fully reusing of non-cloud infrastructure, reduce equipment cost, remarkable in economical benefits.Solve existing
The unified management scheduling problem of the hardware of cloud do not supported by net.Further, resource United Dispatching framework with the operation of existing cloudization is
System, virtualization system and hardware management system compatible.The user that performance requirement is high can be directly deployed on PM, it is not necessary to worry by
This problem of management brought or be deployed in the performance loss of VM.Additionally, scheduling controller grasps user profile, configurable multiple
Dispatching algorithm.
Although combine each embodiment invention has been described at this, but the right of the present invention can not be limited with this
Scope, is implementing during the present invention for required protection, those skilled in the art by check described accompanying drawing, disclosure,
And appended claims, it will be appreciated that and realize all or part of flow process of above-described embodiment, and according to the claims in the present invention
The equivalent variations made, still falls within the scope that invention is contained.In the claims, " include " that a word is not excluded for other composition portions
Dividing or step, "a" or "an" is not excluded for multiple situations.Single controller or other unit can realize in claim
Some the functions enumerated.Some wording that mutually different dependent claims is described, it is not intended that these wording
Can not combine and produce good effect.Computer program can be stored/distributed in suitable medium, such as: optical storage
Medium or solid state medium, provide or as the part of hardware, it would however also be possible to employ other distribution modes together with other hardware, as
(English: Internet) or other wired or wireless systems by the Internet.
Claims (12)
1. resource United Dispatching framework under an isomery cloud environment, it is characterised in that including: scheduling controller, at least one meter
Operator node and each self-corresponding request agent node of described calculating node, described scheduling controller is with described request agent node even
Connecing, described request agent node is connected with described calculating node；Described scheduling controller send for receiving terminal apparatus the
One service request, and ask to described request agent node transmission second service according to the request of described first service, described first
It is that the distribution of described terminal unit calculates node that service request is used for asking described scheduling controller, and the request of described second service is used for
The calculating node of instruction different operating system provides, for described terminal unit, the service of calculating；Described request agent node is for by institute
State second service ask the service command being converted in different operating system and send described service command to different operating system
The calculating node that system is corresponding, it is that described terminal unit provides the service of calculating that described calculating node is used for according to described service command,
Described operating system includes cloud operating system and non-cloud operating system.
Framework the most according to claim 1, it is characterised in that described framework also include manager, at least one control joint
Point and each self-corresponding control agent node of described control node, described manager is for receiving the first management that manager sends
Request, described first management request is used for asking described manager to be managed described calculating node；Described manager is also used
In sending the second management request according to described first management request to described control agent node, described second management request is used for
Indicate described control node that described calculating node is managed；Described control agent node is for by described second management request
Be converted to the administration order in different operating system and by control joint corresponding for described administration order transmission to different operating system
Point, described control node is for being managed described calculating node according to described administration order.
Framework the most according to claim 2, it is characterised in that described control node is additionally operable to performing described management life
Sending the first execution result to described control agent node after order, described control agent node is additionally operable to perform knot by described first
Fruit reports described manager, and described manager is additionally operable to perform result according to described first and updates control node state table.
4. according to the framework described in Claims 2 or 3, it is characterised in that target control joint is carried in described second management request
The mark of point, described second management request is converted to symbol for described control agent node by the mark of described target control node
Close the administration order of the command format of operating system corresponding to described target control node, and by right for described target control node institute
The administration order answering the command format of operating system sends to described target control node.
5. according to the method described in any one of Claims 1-4, it is characterised in that described calculating node is additionally operable to performing
Sending the second execution result to described request agent node after described service command, described request agent node is additionally operable to described
Second performs result reports described scheduling controller, and described scheduling controller is additionally operable to perform result according to described second and updates
Calculate node state table.
6. according to the framework described in any one of claim 1 to 5, it is characterised in that target is carried in the request of described second service
Calculating the mark of node, the request of described second service is turned by the mark of described target computing nodes for described request agent node
It is changed to meet the service command of the command format of operating system corresponding to described target computing nodes, and described target is calculated joint
The service command of the command format of the corresponding operating system of point sends to described target computing nodes.
7. resource United Dispatching method under an isomery cloud environment, it is characterised in that including:
The first service request that scheduling controller receiving terminal apparatus sends, the request of described first service is used for asking described scheduling
Controller is that the distribution of described terminal unit calculates node；
Described scheduling controller sends second service request according to the request of described first service to destination request agent node, described
Second service request is used for indicating target computing nodes to provide, for described terminal unit, the service of calculating；
The request of described second service is converted to meet and grasps corresponding to described target computing nodes by described destination request agent node
Make the service command of the command format of system；
Described service command is sent to described target computing nodes by described destination request agent node；
Described target computing nodes provides, for described terminal unit, the service of calculating according to described service command.
Method the most according to claim 7, it is characterised in that the first of described scheduling controller receiving terminal apparatus transmission
Before service request, also include:
Manager receives the first management request, carries the mark of described target computing nodes in described first management request, described
First management request is used for asking described manager to start described target computing nodes；
Described manager sends the second management according to described first management request to target control agent node and asks, and described second
Management request is used for indicating target control node to start described target computing nodes；
Described second management request is converted to meet and grasps corresponding to described target control node by described target control agent node
Make the administration order of the command format of system；
Described administration order is sent to described target control node by described target control agent node；
Described target control node starts described target computing nodes according to described administration order.
Method the most according to claim 8, it is characterised in that described target control node starts according to described administration order
After described target computing nodes, also include:
Described target control node sends the first execution result to described target control agent node；
Described target control agent node performs result by described first and is sent to described manager；
Described manager performs result according to described first and updates control node state table, and described control node state table is used for institute
State manager from least one control agent node, determine described target control agent node.
Method the most according to claim 8 or claim 9, it is characterised in that carry described target control in described second management request
The mark of node processed, for described target control agent node, the mark of described target control node determines that described target control saves
Point.
11. according to the method described in any one of claim 7 to 10, it is characterised in that described target computing nodes is according to described
After service command provides, for described terminal unit, the service of calculating, also include:
Described target computing nodes sends the second execution result to described request agent node；
Described request agent node performs result by described second and is sent to described scheduling controller；
Described scheduling controller performs result according to described second and updates calculating node state table, and described calculating node state table is used
From at least one calculating node, described target computing nodes is determined in described scheduling controller.
12. according to the method described in any one of claim 7 or 11, it is characterised in that carry institute in the request of described second service
State the mark of target computing nodes, for described destination request agent node, described target computing nodes determines that described target calculates
Node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610644721.5A CN106254471A (en) | 2016-08-09 | 2016-08-09 | Resource United Dispatching method and system under a kind of isomery cloud environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610644721.5A CN106254471A (en) | 2016-08-09 | 2016-08-09 | Resource United Dispatching method and system under a kind of isomery cloud environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106254471A true CN106254471A (en) | 2016-12-21 |
Family
ID=58078764
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610644721.5A Pending CN106254471A (en) | 2016-08-09 | 2016-08-09 | Resource United Dispatching method and system under a kind of isomery cloud environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106254471A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106713507A (en) * | 2017-02-22 | 2017-05-24 | 上海斐讯数据通信技术有限公司 | Management method and management system for batches of cloud terminal devices |
| CN106775953A (en) * | 2016-12-30 | 2017-05-31 | 北京中电普华信息技术有限公司 | Realize the method and system of OpenStack High Availabitities |
| CN106802805A (en) * | 2017-01-18 | 2017-06-06 | 郑州云海信息技术有限公司 | A kind of application service management method and device for being applicable server admin |
| CN107239290A (en) * | 2017-06-12 | 2017-10-10 | 郑州云海信息技术有限公司 | The management method and device of a kind of application service |
| CN107358105A (en) * | 2017-06-01 | 2017-11-17 | 上海红阵信息科技有限公司 | Isomery function equivalence body phase opposite sex measuring method, distribution method, device and equipment |
| CN108989384A (en) * | 2018-05-31 | 2018-12-11 | 华为技术有限公司 | A kind of method of data processing, cloudy management system and relevant device |
| CN110245170A (en) * | 2019-04-19 | 2019-09-17 | 联通***集成有限公司 | Data processing method and system |
| CN111164945A (en) * | 2017-10-13 | 2020-05-15 | 瑞典爱立信有限公司 | Method and apparatus for brokering between different architectures |
| CN112953993A (en) * | 2019-12-11 | 2021-06-11 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, network system and storage medium |
| WO2021121067A1 (en) * | 2019-12-20 | 2021-06-24 | 深圳前海微众银行股份有限公司 | Task execution method and apparatus |
| CN113254220A (en) * | 2021-07-01 | 2021-08-13 | 国汽智控(北京)科技有限公司 | Networked automobile load cooperative control method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102014159A (en) * | 2010-11-29 | 2011-04-13 | 华中科技大学 | Layered resource reservation system under cloud computing environment |
| CN102655532A (en) * | 2012-04-18 | 2012-09-05 | 上海和辰信息技术有限公司 | Distributed heterogeneous virtual resource integration management method and system |
| CN102739771A (en) * | 2012-04-18 | 2012-10-17 | 上海和辰信息技术有限公司 | Cloud application integrated management platform and method supporting service fusion |
| CN104063560A (en) * | 2014-07-08 | 2014-09-24 | 广东轩辕网络科技股份有限公司 | Dispatching system and method based on cloud computing platform |
-
2016
- 2016-08-09 CN CN201610644721.5A patent/CN106254471A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102014159A (en) * | 2010-11-29 | 2011-04-13 | 华中科技大学 | Layered resource reservation system under cloud computing environment |
| CN102655532A (en) * | 2012-04-18 | 2012-09-05 | 上海和辰信息技术有限公司 | Distributed heterogeneous virtual resource integration management method and system |
| CN102739771A (en) * | 2012-04-18 | 2012-10-17 | 上海和辰信息技术有限公司 | Cloud application integrated management platform and method supporting service fusion |
| CN104063560A (en) * | 2014-07-08 | 2014-09-24 | 广东轩辕网络科技股份有限公司 | Dispatching system and method based on cloud computing platform |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106775953A (en) * | 2016-12-30 | 2017-05-31 | 北京中电普华信息技术有限公司 | Realize the method and system of OpenStack High Availabitities |
| CN106802805A (en) * | 2017-01-18 | 2017-06-06 | 郑州云海信息技术有限公司 | A kind of application service management method and device for being applicable server admin |
| CN106713507B (en) * | 2017-02-22 | 2020-01-14 | 上海斐讯数据通信技术有限公司 | Management method and management system for batch cloud terminal equipment |
| CN106713507A (en) * | 2017-02-22 | 2017-05-24 | 上海斐讯数据通信技术有限公司 | Management method and management system for batches of cloud terminal devices |
| CN107358105A (en) * | 2017-06-01 | 2017-11-17 | 上海红阵信息科技有限公司 | Isomery function equivalence body phase opposite sex measuring method, distribution method, device and equipment |
| CN107358105B (en) * | 2017-06-01 | 2020-05-29 | 珠海高凌信息科技股份有限公司 | Method, device and equipment for measuring and distributing isomerism functional equivalent |
| CN107239290A (en) * | 2017-06-12 | 2017-10-10 | 郑州云海信息技术有限公司 | The management method and device of a kind of application service |
| CN111164945A (en) * | 2017-10-13 | 2020-05-15 | 瑞典爱立信有限公司 | Method and apparatus for brokering between different architectures |
| US11750708B2 (en) | 2017-10-13 | 2023-09-05 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and device for proxy between different architectures |
| CN108989384A (en) * | 2018-05-31 | 2018-12-11 | 华为技术有限公司 | A kind of method of data processing, cloudy management system and relevant device |
| CN108989384B (en) * | 2018-05-31 | 2020-08-14 | 华为技术有限公司 | Data processing method, multi-cloud management system and related equipment |
| US11765243B2 (en) | 2018-05-31 | 2023-09-19 | Huawei Technologies Co., Ltd. | Data processing method, multi-cloud management system, and related device |
| CN110245170A (en) * | 2019-04-19 | 2019-09-17 | 联通***集成有限公司 | Data processing method and system |
| CN110245170B (en) * | 2019-04-19 | 2021-11-16 | 联通数字科技有限公司 | Data processing method and system |
| CN112953993A (en) * | 2019-12-11 | 2021-06-11 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, network system and storage medium |
| CN112953993B (en) * | 2019-12-11 | 2023-09-26 | 阿里巴巴集团控股有限公司 | Resource scheduling method, equipment, network system and storage medium |
| WO2021121067A1 (en) * | 2019-12-20 | 2021-06-24 | 深圳前海微众银行股份有限公司 | Task execution method and apparatus |
| CN113254220A (en) * | 2021-07-01 | 2021-08-13 | 国汽智控(北京)科技有限公司 | Networked automobile load cooperative control method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106254471A (en) | Resource United Dispatching method and system under a kind of isomery cloud environment | |
| US9967136B2 (en) | System and method for policy-based smart placement for network function virtualization | |
| CN109743415B (en) | Public cloud network elastic IP implementation method and system | |
| US20200084107A1 (en) | Method And Apparatus For Creating Network Slice, And Communications System | |
| CN105812171B (en) | A kind of the network function VNF control method and equipment of virtualization | |
| CN109669758A (en) | Concocting method, device, equipment and the storage medium of server resource | |
| CN104850450A (en) | Load balancing method and system facing mixed cloud application | |
| CN103051564B (en) | The method and apparatus of dynamic resource allocation | |
| US20150156131A1 (en) | Method and system of geographic migration of workloads between private and public clouds | |
| CN113742031B (en) | Node state information acquisition method and device, electronic equipment and readable storage medium | |
| CN107534570A (en) | Virtualize network function monitoring | |
| CN105138389A (en) | Method and system for managing virtual devices in cluster | |
| CN110661842B (en) | Resource scheduling management method, electronic equipment and storage medium | |
| CN104246741A (en) | Orchestrating hybrid cloud services | |
| CN109669762A (en) | Cloud computing resources management method, device, equipment and computer readable storage medium | |
| CN109257399B (en) | Cloud platform application program management method, management platform and storage medium | |
| JP2016103179A (en) | Allocation method for computer resource and computer system | |
| CN103873534A (en) | Method and device for application cluster migration | |
| CN104601680A (en) | Resource management method and device | |
| CN106155811A (en) | Graphic processing facility, resource service device, resource regulating method and device | |
| CN106293933A (en) | A kind of cluster resource configuration supporting much data Computational frames and dispatching method | |
| CN103780428A (en) | Centralized resource management method and system applied to cloud architecture | |
| CN107920117A (en) | A kind of method for managing resource, control device and resource management system | |
| CN109960579B (en) | Method and device for adjusting service container | |
| CN105404530A (en) | System and method for realizing simple deployment and using private cloud |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161221 |
|
| RJ01 | Rejection of invention patent application after publication |