CN102567072B - Resource allocation method, resource allocation device and resource allocation system - Google Patents
Resource allocation method, resource allocation device and resource allocation system Download PDFInfo
- Publication number
- CN102567072B CN102567072B CN201010597688.8A CN201010597688A CN102567072B CN 102567072 B CN102567072 B CN 102567072B CN 201010597688 A CN201010597688 A CN 201010597688A CN 102567072 B CN102567072 B CN 102567072B
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- physical resource
- priority
- distribution priority
- performance information
- 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.)
- Active
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a resource allocation method, a resource allocation device and a resource allocation system. The resource allocation method includes the steps: utilizing a virtual machine manager to receive virtualized business characteristic information on a first server of current handling business reported respectively by all virtual machines; and allocating physical resources to the virtual machines according to the business characteristic information. By the aid of the resource allocation method adopting the scheme in the embodiment, resource allocation for the virtual machines is more reasonable, and service quality of the virtual machines for business handling is improved.
Description
Technical field
The present invention relates to the technical field of virtualization in field of computer technology, particularly relate to a kind of resource allocation methods of virtual machine, Apparatus and system.
Background technology
Share to realize the virtual of computer physics resource better, fully improve the utilization factor of computer physics resource, reach the object such as energy-saving and emission-reduction and load balancing, Intel Virtualization Technology is more and more subject to the attention of IT and association area, and obtains a wide range of applications.
Server virtualization refers to and utilizes Intel Virtualization Technology that a physical server is virtualized into multiple virtual machine, each virtual machine can independently provide the function that same physical server is identical, as: installing operating system, database and disposing application program etc. on a virtual machine.Like this, different virtual machine just can encapsulate different service application logics, namely on same physical computer, runs multiple independently operating system and separate application.And the physical resource that server has (as CPU, internal memory, disk and Internet resources etc.) can be shared by these virtual machines.
In current server virtualization technology application, primarily of the resource consumption situation of virtual machine manager according to the physical resource layer of each virtual machine on server, relative strategy is adopted to complete the assigning process of server physical resource (as CPU, internal memory, disk and Internet resources etc.).
The present inventor states in technical scheme on the implementation and finds, it is only the distribution carrying out physical resource according to the resource consumption situation of physical resource, for each virtual machine distributes corresponding physical resource, often because Resourse Distribute is reasonable not, and there is the situation that cannot meet the resource requirement of partial virtual machine when the business of process, cause the service quality of virtual machine to business processing lower.
Summary of the invention
The embodiment of the present invention provides a kind of resource allocation methods, Apparatus and system, in order to realize for virtual machine more reasonably Resourse Distribute, improves the service quality that virtual machine carries out business processing.
The embodiment of the present invention provides a kind of resource allocation methods, comprising:
Virtual machine manager receives the traffic performance information of the current process business that virtualized each virtual machine reports respectively in first server; And
Be that described each virtual machine distributes physical resource according to described traffic performance information.
The embodiment of the present invention also provides a kind of virtual machine manager, comprising:
Receiving element, for receiving the traffic performance information of the current process business that virtualized each virtual machine in first server reports respectively;
Allocation units, for according to described traffic performance information being described each virtual machine distribution physical resource.
The embodiment of the present invention also provides a kind of virtual machine, comprising:
Collecting unit, for gathering the traffic performance information of the current process business of this virtual machine;
Transmitting element, for by described traffic performance information reporting to virtual machine manager.
The embodiment of the present invention also provides a kind of resource allocation system, comprising: virtual machine and virtual machine manager, wherein:
Described virtual machine, for gathering the traffic performance information of the current process business of this virtual machine; And give described virtual machine manager by described traffic performance information reporting;
Described virtual machine manager, the described traffic performance information for reporting according to the described each virtual machine received is that described each virtual machine distributes physical resource.
In the method that the embodiment of the present invention provides, on server, virtualized each virtual machine gathers the traffic performance information of its current process business, and by the traffic performance information reporting of collection to virtual machine manager; The traffic performance information that virtual machine manager then reports respectively according to each virtual machine received is that each virtual machine distributes physical resource.The traffic performance information reported due to virtual machine can characterize this virtual machine when the inherent characteristic of the business of pre-treatment and the current situation to this business processing, and then make virtual machine manager can for this traffic performance information, the carrying out of the physical resource of each virtual machine place server is distributed, realize more reasonably Resourse Distribute, also just achieving for each virtual machine current process business is that each virtual machine distributes physical resource, can business demand be better met, improve the service quality that virtual machine carries out business processing.
Accompanying drawing explanation
The process flow diagram of the resource allocation methods that Fig. 1 provides for the embodiment of the present invention;
The detail flowchart of the resource allocation methods that Fig. 2 provides for the embodiment of the present invention;
The structural representation of the virtual machine manager that Fig. 3 provides for the embodiment of the present invention;
The structural representation of the allocation units in the virtual machine manager that Fig. 4 provides for the embodiment of the present invention;
The structural representation of the virtual machine that Fig. 5 provides for the embodiment of the present invention;
The structural representation of the resource allocation system that Fig. 6 provides for the embodiment of the present invention.
Embodiment
More reasonably Resourse Distribute is realized in order to provide, improve the implementation that virtual machine carries out the service quality of business processing, embodiments provide a kind of resource allocation methods, Apparatus and system, below in conjunction with Figure of description, the preferred embodiments of the present invention are described, be to be understood that, preferred embodiment described herein, only for instruction and explanation of the present invention, is not intended to limit the present invention.And when not conflicting, the embodiment in the application and the feature in embodiment can combine mutually.
The embodiment of the present invention provides a kind of resource allocation methods, as shown in Figure 1, comprising:
Step S101, virtual machine manager receive the traffic performance information of the current process business that virtualized each virtual machine reports respectively in first server.
Step S102, be that each virtual machine distributes physical resource according to this traffic performance information.
In above-mentioned steps S102, first can analyze this traffic performance information received, determine the resource allocation policy for each virtual machine, and be the distribution that each virtual machine carries out physical resource according to the resource allocation policy determined.
Below in conjunction with accompanying drawing, with specific embodiment, method provided by the invention and device and corresponding system are described in detail.
Figure 2 shows that the detail flowchart of the resource allocation methods that the embodiment of the present invention provides, specifically comprise:
The collection of step S201, traffic performance information with report.
Virtual machine is after startup optimization, determine that its current encapsulation business is its current process business, gather the traffic performance information of its current process business, this traffic performance information is specifically as follows the relevant various information of process business current to it, as characterized the information of this virtual machine when the inherent characteristic of the business of pre-treatment, and the information of the current situation to this business processing, such as, can for service identification, process this business time the business process mark etc. of each business process that generates.
After collecting traffic performance information, reported virtual machine manager.
In this step business customizing messages collection and on call time, can be periodic triggering, also can be setting event trigger, as in the starting stage of carrying out business processing, first reporting service mark, in the interstage of business processing, reports the business process of each business process to identify.
The analysis of step S202, traffic performance information.
Virtual machine manager, after receiving the traffic performance information that each virtual machine reports, is analyzed it, determines follow-up further information required when determining resource allocation policy, such as:
According to the service identification in traffic performance information, inquiry business mark and the corresponding relation of type of service, determine the type of service corresponding to service identification, the type of service of this type of service and the current process business of virtual machine.
Further according to the service quality of business in reality and demand characteristics, can also pre-set the priority of often kind of type of service, virtual machine manager then according to the type of service determined, can inquire the priority of this type of service.
According to the demand characteristics of each business process generated when processing business in reality, can also pre-set the priority of miscellaneous service process, virtual machine manager then according to the business process mark reported, can determine the priority corresponding to business process mark.
Step S203, determine resource allocation policy.
Virtual machine manager based in above-mentioned steps S202 to the analysis of the respective traffic performance information that each virtual machine reports respectively, determine resource allocation policy, the determination of resource allocation policy can be considered around objects such as QoS, energy-saving and emission-reduction, load balancing and resource optimizations, specifically can consider to determine corresponding resource allocation policy from many aspects, as follows:
First kind of way: consider the business that in reality, type of service is different, its business demand characteristic may be different, such as: delay requirement, network bandwidth requirements, business real-time demand, data storage requirement etc., namely type of service can indicate the business demand characteristic of specifying physical resource for certain, such as, real-time class business is higher to delay requirement, so higher to the demand of cpu resource, content class service is higher to data storage requirement, so more high to the demand of storage resources.
So, the embodiment of the present invention proposes, virtual machine manager can type of service corresponding to the service identification in traffic performance information, determine the distribution priority of each virtual machine when distributing appointment physical resource for each virtual machine, and determine that resource allocation policy is, according to the distribution priority of each virtual machine, specify physical resource for each virtual machine distributes; Wherein, the determination distributing priority is specific as follows:
When specifying physical resource to be cpu resource, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of real-time class business, is the distribution priority of the virtual machine of non real-time class business higher than the type of service corresponding to the service identification reported;
When specifying physical resource to be storage resources, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of content class service, is the distribution priority of the virtual machine of non-content class business higher than the type of service corresponding to the service identification reported;
For the business of other types, business demand characteristic that in like manner can be concrete according to it, make and specify distribution priority during physical source distributing to determine rule for other certain, and by virtual machine manager according to this rule, when specifying physical source distributing for other certain, determine the distribution priority of each virtual machine, no longer illustrating at this is described in detail.
The second way: virtual machine manager can also the priority of type of service corresponding to the service identification in traffic performance information, determine the distribution priority of each virtual machine, and determine that resource allocation policy is, be that each virtual machine distributes physical resource according to the distribution priority of each virtual machine.
Specifically can preset distribution priority really to establish rules, be, the distribution priority of the higher then corresponding virtual machine of priority of type of service is higher, so, virtual machine manager is according to this rule, determine the distribution priority of the virtual machine that the priority of the type of service corresponding to service identification reported is high, the virtual machine that the priority higher than the type of service corresponding to the service identification reported is low.
The third mode: the determination mode of the distribution priority of the virtual machine that above-mentioned first kind of way and the second way are proposed, when the type of service corresponding to the service identification in the traffic performance information that different virtual machine reports respectively is identical, the distribution priority then determined this different virtual machine is identical, even if consider that type of service is identical, but the business that virtual machine works as pre-treatment may be different, so, the embodiment of the present invention also proposed for the identical each virtual machine of the type of service corresponding to the service identification reported, the further scheme distinguishing distribution priority, specific as follows:
Mode A: the business process generated when considering and process different business is different, so, propose based on business process information, further differentiation distributes priority, specifically can preset the priority of each business process, the distribution priority that distribution priority is established rules really then for the higher then corresponding virtual machine of priority of business process is higher.Virtual machine manager is then according to this rule, determine each virtual machine that the type of service corresponding to service identification that reports is identical, according to the priority corresponding to the business process mark in the traffic performance information that each virtual machine determined reports respectively, the distribution priority of the virtual machine that the priority corresponding to business process mark determining to report is high, higher than the distribution priority of the low virtual machine of the priority corresponding to the business process mark reported.
In the embodiment of the present invention, virtual machine can only report the business process of part main business process to identify, and is only identified according to the business process of this main steps by virtual machine manager, distinguishes further and distributes priority.
Mode B: propose the Current resource utilization factor based on each virtual machine, distinguish further and distribute priority, concrete distribution priority is established rules really, is, the distribution priority of the higher then corresponding virtual machine of Current resource utilization factor is higher.Virtual machine manager is then according to this rule, determine each virtual machine that the type of service corresponding to service identification that reports is identical, according to the Current resource utilization factor of each virtual machine determined, determine the distribution priority of the virtual machine that Current resource utilization factor is high, higher than the distribution priority of the low virtual machine of Current resource utilization factor.
Wherein, the Current resource utilization factor of virtual machine by virtual machine active reporting to Virtual Machine Management Service device, also can be monitored the resource service condition of each virtual machine, determine the Current resource utilization factor of each virtual machine by Virtual Machine Management Service device.
In the embodiment of the present invention, (the distribution priority for each virtual machine during appointment physical source distributing is comprised) after the distribution priority determining each virtual machine, then determine that resource allocation policy is, be that each virtual machine distributes physical resource (comprise specify physical resource) according to the distribution priority of each virtual machine, namely can according to distribution priority order from high in the end, for each virtual machine distributes physical resource, the preferential physical resource demand meeting the high virtual machine of distribution priority; Specifically can distribute in conjunction with the Current resource utilization factor of each virtual machine, as follows:
First preset the setting resource utilization threshold that each virtual machine is corresponding, according to Current resource utilization factor and the corresponding magnitude relationship setting resource utilization threshold of each virtual machine, determine the physical resource amount of the current needs of each virtual machine.
Such as, Current resource utilization factor exceeds setting resource utilization threshold, represent and need to increase the physical resource distributed, the concrete physical resource amount needing to increase, can determine according to the difference between Current resource utilization factor and setting resource utilization threshold, meet after increasing the physical resource distributed, resource utilization does not exceed setting resource utilization threshold;
Current resource utilization factor does not exceed setting resource utilization threshold, represent and do not need to increase the physical resource distributed, and, can also according to the difference between Current resource utilization factor and setting resource utilization threshold, the unnecessary physics stock number that can provide is provided, meet after subtracting under absorbed physical resource, resource utilization does not exceed setting resource utilization threshold.
After the physical resource amount determining the current needs of each virtual machine, namely can need physical resource amount according to current, determine the physical resource amount distributing to each virtual machine, determine that rule meets the physical resource demand of distributing the high virtual machine of priority for preferential.During concrete distribution, namely virtual machine manager according to distribution priority order from high to low, for needing the virtual machine increasing physical resource, can determine that it needs the physical resource amount increased; For not needing the virtual machine increasing physical resource, the physical resource amount of distribution can be kept constant; When the remaining physical resource of server is inadequate, need the physical resource reducing virtual machine, to be supplied to other when needing the virtual machine increasing physical resource, according to distribution priority order from low to high, reduce the physical resource that virtual machine distributes.
In the embodiment of the present invention, for the setting of above-mentioned setting resource utilization, propose to distribute setting resource utilization threshold corresponding to the high virtual machine of priority higher than the setting resource utilization threshold distributing the low virtual machine of priority.
Consider that the physical resource that a physical server can provide is limited, the situation of physical resource deficiency may be there is, the embodiment of the present invention proposes, when the traffic performance information according to each virtual machine, determine the physical resource needed for each virtual machine distributes, when exceeding the physical resource that this first server can distribute, in each virtual machine, select part virtual machine (vm) migration on second server, and be the resource allocation policy that the virtual machine do not moved distributes physical resource based on the physical resource that first server can be distributed; And the resource allocation policy of physical resource is distributed based on the virtual machine that the physical resource that second server can distribute is migration.
Preferably, when determining whether migration, whether it is also conceivable to the current distribution priority that cannot meet the virtual machine of physical resource demand higher than setting priority, if higher than, then move, otherwise do not move.
Specifically determine the virtual machine moved, following scheme can be adopted:
According to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server; Or
The virtual machine that the physical resource selecting needs to distribute remains unchanged also migrates on second server; Or
According to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server in each virtual machine needing the physical resource distributed to remain unchanged.
In this step, the resource allocation policy determination rale store that virtual machine manager can preset is in allocation strategy storehouse, when determining resource allocation policy, first from allocation strategy storehouse, Gains resources allocation strategy determines rule, and determines corresponding resource allocation policy according to traffic performance information according to this rule.
Step S204, carry out physical source distributing.
Virtual machine manager, after determining the resource allocation policy for each virtual machine, namely carries out the distribution of physical resource according to this resource allocation policy.
After the distribution carrying out physical resource, record can also be carried out to resource allocation conditions, for follow-up relevant Resourse Distribute provides corresponding reference data.
Understand the above-mentioned resource allocation methods that the embodiment of the present invention provides in order to be more convenient for, be described below by instantiation:
Suppose to deploy three virtual machines altogether on the server of an applying virtual technology, three virtual machines are respectively VoIP business, Streaming business and instant messaging when the business of pre-treatment.Known by the analysis of the traffic performance information to three virtual machine current process business, VoIP belongs to real-time service, requires higher, can be set as high-grade business to time delay and treatment effeciency etc.; Streaming business belongs to content class service, data are stored and mutual requirement higher, but requirement of real-time is not high, can be set as ordinary grade business; Instant messaging due to its interaction data amount relatively less, its requirement of real-time is lower, can be set as inferior grade business.
According to the real time characteristic of above-mentioned three kinds of business, determine the resource allocation policy for cpu resource, such as:
In the starting stage of business processing, can determine the distribution priority of each virtual machine according to the priority of each type of service, namely the priority of type of service is higher, then the distribution priority of corresponding virtual machine is higher;
In the interstage of business processing, suppose that the cpu resource of the current virtual machine distribution for process VoIP business, Streaming business and instant messaging is respectively: 50%, 20% and 15%; And for VoIP business, because the priority of itself is higher, and determine that the utilization rate of its current C PU resource has exceeded corresponding setting threshold value, so the distribution priority of the virtual machine of process VoIP business is defined as limit priority; For Streaming business and instant messaging, the utilization rate of its cpu resource separately does not all exceed corresponding setting threshold value, but because the priority of Streaming business is higher than the priority of instant messaging, so determine the distribution priority of the distribution priority of the virtual machine for the treatment of S treaming business higher than the virtual machine of process instant message.
According to distribution priority order from high to low, be preferably and distribute the high virtual machine of priority and distribute cpu resource required for it, or distribute shared total cpu resource ratio higher cpu resource for distributing the high virtual machine of priority.Such as: the cpu resource that the virtual machine for process VoIP business distributes is adjusted to 55%, and the cpu resource that the virtual machine remaining treatment S treaming business distributes is constant, and the cpu resource that the virtual machine for processing instant messaging distributes is adjusted to 15%.
Based on same inventive concept, according to the resource allocation methods that the above embodiment of the present invention provides, correspondingly, another embodiment of the present invention additionally provides a kind of virtual machine manager, and its structural representation as shown in Figure 3, specifically comprises:
Receiving element 301, for receiving the traffic performance information of the current process business that virtualized each virtual machine in first server reports respectively;
Allocation units 302, for according to this traffic performance information being each virtual machine distribution physical resource.
Preferably, allocation units 302, specifically for the type of service corresponding to the service identification in the traffic performance information that reports respectively according to each virtual machine, determine being the distribution priority that each virtual machine distributes when specifying physical resource; And according to the distribution priority determined, for each virtual machine distributes this appointment physical resource.
Preferably, allocation units 302, specifically for when specifying physical resource to be cpu resource, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of real-time class business, is the distribution priority of the virtual machine of non real-time class business higher than the type of service corresponding to the service identification reported; And/or
When specifying physical resource to be storage resources, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of content class service, is the distribution priority of the virtual machine of non-content class business higher than the type of service corresponding to the service identification reported.
Preferably, allocation units 302, specifically for the priority of the type of service corresponding to the service identification in the traffic performance information that reports respectively according to each virtual machine, determine the distribution priority of the virtual machine that the priority of the type of service corresponding to service identification reported is high, the virtual machine that the priority higher than the type of service corresponding to the service identification reported is low; And according to the distribution priority determined, for each virtual machine distributes physical resource.
Preferably, allocation units 302, also for according to the distribution priority determined, before described each virtual machine distributes physical resource, determine each virtual machine that the type of service corresponding to service identification that reports is identical;
According to the priority corresponding to the business process mark in the traffic performance information that each virtual machine determined reports respectively, the distribution priority of the virtual machine that the priority corresponding to business process mark determining to report is high, higher than the distribution priority of the low virtual machine of the priority corresponding to the business process mark reported; Or
According to the Current resource utilization factor of each virtual machine determined, determine the distribution priority of the virtual machine that Current resource utilization factor is high, higher than the distribution priority of the low virtual machine of Current resource utilization factor.
Preferably, allocation units 302, specifically for according to the Current resource utilization factor of each virtual machine and the corresponding magnitude relationship setting resource utilization threshold, determine the physical resource amount of the current needs of each virtual machine; And according to the physical resource amount of the current needs of each virtual machine, based on distribution priority order from high to low, be that each virtual machine distributes physical resource.
Preferably, allocation units 302, specifically for when determining according to the traffic performance information of each virtual machine the physical resource needed for each virtual machine distributes, when exceeding the physical resource that first server can distribute, in each virtual machine, select part virtual machine (vm) migration on second server; And be that the virtual machine do not moved distributes physical resource based on the physical resource that first server can be distributed; And distribute physical resource based on the virtual machine that the physical resource that second server can distribute is migration.
Preferably, described allocation units 302, specifically for according to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server; Or
The virtual machine that the physical resource selecting needs to distribute remains unchanged also migrates on second server; Or
According to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server in each virtual machine needing the physical resource distributed to remain unchanged.
In the embodiment of the present invention, above-mentioned virtual machine manager is the core in virtual management system, except completing except the operation of above-mentioned functions by its allocation units 302, also the function such as life cycle management (establishment, deletion, deployment etc.), real-time migration, virtual unit hot plug of virtual machine will be completed.Wherein, as shown in Figure 4, allocation units 302 further can also be divided into traffic performance information analysis subelement 401 and Resourse Distribute subelement 402, traffic performance information analysis subelement 401 is analyzed for the traffic performance information reported the virtual machine received, determine allocation strategy, and distribute subelement 402 by the allocation strategy announcing resource determined, carry out physical source distributing by Resourse Distribute subelement 402 according to this allocation strategy, for virtual machine distributes physical resource; Concrete, allocation strategy storehouse 303 can also be comprised, rule is determined for storing the resource allocation policy pre-established, for traffic performance information analysis subelement 401 when determining allocation strategy, policy library interface 4011 and allocation strategy interface 4012 can be comprised in traffic performance information analysis subelement 401, be respectively used to carry out alternately with allocation strategy storehouse 303 and Resourse Distribute subelement 402.
Based on same inventive concept, according to the resource allocation methods that the above embodiment of the present invention provides, correspondingly, another embodiment of the present invention additionally provides a kind of virtual machine, and its structural representation as shown in Figure 5, specifically comprises:
Collecting unit 501, for gathering the traffic performance information of the current process business of this virtual machine;
Transmitting element 502, for will gather traffic performance information reporting to virtual machine manager.
In the embodiment of the present invention, collecting unit 501 can be deployed in each virtual machine internal, run with kernel process form.In order to ensure the normal operation of collecting unit 501, when virtual machine activation, by reading and running the relevant configuration information of collecting unit 501, start collecting unit 501.Concrete, preset configuration file can be adopted, and this configuration file is added virtual machine VBD list with CD-ROM drive form, this CD-ROM drive can be scanned when virtual machine activation carries out hardware scanning, and load relevant information, complete the startup of collecting unit 501.
Based on same inventive concept, according to the resource allocation methods that the above embodiment of the present invention provides, correspondingly, another embodiment of the present invention additionally provides a kind of resource allocation system, its structural representation as shown in Figure 6, specifically comprises: virtual machine 601 and virtual machine manager 602, wherein:
Virtual machine 601, for gathering the traffic performance information of the current process business of this virtual machine; And by traffic performance information reporting to virtual machine manager 602;
Virtual machine manager 602 is that each virtual machine 601 distributes physical resource for the traffic performance information reported according to each virtual machine received.
In sum, the scheme that the embodiment of the present invention provides, comprising: virtual machine manager receives the traffic performance information of the current process business that virtualized each virtual machine reports respectively in first server; And be that each virtual machine distributes physical resource according to this traffic performance information.The scheme adopting the embodiment of the present invention to provide, achieves the more reasonably Resourse Distribute for virtual machine, improves the service quality that virtual machine carries out business processing.
Obviously, those skilled in the art can carry out various change and modification to the present invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.
Claims (16)
1. a resource allocation methods, is characterized in that, comprising:
Virtual machine manager receives the traffic performance information of the current process business that virtualized each virtual machine reports respectively in first server; And
Be that described each virtual machine distributes physical resource according to described traffic performance information;
Wherein, be that described each virtual machine distributes physical resource according to described traffic performance information, specifically comprise: according to the type of service corresponding to the service identification in the traffic performance information that each virtual machine reports respectively, determine the distribution priority when distributing appointment physical resource for described each virtual machine; And according to the distribution priority determined, for described each virtual machine distributes described appointment physical resource.
2. the method for claim 1, is characterized in that, according to the type of service corresponding to the service identification in the traffic performance information that each virtual machine reports respectively, determines the distribution priority when distributing appointment physical resource for described each virtual machine, specifically comprises:
When described appointment physical resource is cpu resource, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of real-time class business, is the distribution priority of the virtual machine of non real-time class business higher than the type of service corresponding to the service identification reported; And/or
When described appointment physical resource is storage resources, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of content class service, is the distribution priority of the virtual machine of non-content class business higher than the type of service corresponding to the service identification reported.
3. the method for claim 1, is characterized in that, is that described each virtual machine distributes physical resource, specifically comprises according to described traffic performance information:
The priority of the type of service corresponding to the service identification in the traffic performance information that each virtual machine reports respectively, determine the distribution priority of the virtual machine that the priority of the type of service corresponding to service identification reported is high, the virtual machine that the priority higher than the type of service corresponding to the service identification reported is low; And
According to the distribution priority determined, for described each virtual machine distributes physical resource.
4. the method as described in as arbitrary in claim 1-3, is characterized in that, according to the distribution priority determined, before described each virtual machine distributes physical resource, also comprises:
Determine each virtual machine that the type of service corresponding to service identification that reports is identical;
According to the priority corresponding to the business process mark in the traffic performance information that each virtual machine determined reports respectively, the distribution priority of the virtual machine that the priority corresponding to business process mark determining to report is high, higher than the distribution priority of the low virtual machine of the priority corresponding to the business process mark reported; Or
According to the Current resource utilization factor of each virtual machine determined, determine the distribution priority of the virtual machine that Current resource utilization factor is high, higher than the distribution priority of the low virtual machine of Current resource utilization factor.
5. the method as described in as arbitrary in claim 1-3, is characterized in that, according to the distribution priority determined, is that described each virtual machine distributes physical resource, specifically comprises:
According to Current resource utilization factor and the corresponding magnitude relationship setting resource utilization threshold of described each virtual machine, determine the physical resource amount of the current needs of described each virtual machine;
According to the physical resource amount of the current needs of each virtual machine determined, based on distribution priority order from high to low, for described each virtual machine distributes physical resource.
6. method as claimed in claim 5, is characterized in that, distributes setting resource utilization threshold corresponding to the high virtual machine of priority higher than the setting resource utilization threshold distributing the low virtual machine of priority.
7. the method as described in as arbitrary in claim 1-3, is characterized in that, is that described each virtual machine distributes physical resource, specifically comprises according to described traffic performance information:
When determining the physical resource needed for described each virtual machine distributes according to described traffic performance information, when exceeding the physical resource that described first server can distribute, in described each virtual machine, select part virtual machine and migrate on second server;
The physical resource that can distribute based on described first server is that the virtual machine do not moved distributes physical resource; And
The physical resource that can distribute based on described second server is that the virtual machine moved distributes physical resource.
8. method as claimed in claim 7, is characterized in that, select part virtual machine and migrate on second server, be specially in described each virtual machine:
According to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server; Or
The virtual machine that the physical resource selecting needs to distribute remains unchanged also migrates on second server; Or
According to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server in each virtual machine needing the physical resource distributed to remain unchanged.
9. a virtual machine manager, is characterized in that, comprising:
Receiving element, for receiving the traffic performance information of the current process business that virtualized each virtual machine in first server reports respectively;
Allocation units, for according to described traffic performance information being described each virtual machine distribution physical resource;
Wherein, described allocation units, specifically for the type of service corresponding to the service identification in the traffic performance information that reports respectively according to each virtual machine, determine being the distribution priority that described each virtual machine distributes when specifying physical resource; And according to the distribution priority determined, for described each virtual machine distributes described appointment physical resource.
10. virtual machine manager as claimed in claim 9, it is characterized in that, described allocation units, specifically for when described appointment physical resource is cpu resource, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of real-time class business, is the distribution priority of the virtual machine of non real-time class business higher than the type of service corresponding to the service identification reported; And/or
When described appointment physical resource is storage resources, determining that the type of service corresponding to service identification reported is the distribution priority of the virtual machine of content class service, is the distribution priority of the virtual machine of non-content class business higher than the type of service corresponding to the service identification reported.
11. virtual machine managers as claimed in claim 9, it is characterized in that, described allocation units, specifically for the priority of the type of service corresponding to the service identification in the traffic performance information that reports respectively according to each virtual machine, determine the distribution priority of the virtual machine that the priority of the type of service corresponding to service identification reported is high, the virtual machine that the priority higher than the type of service corresponding to the service identification reported is low; And according to the distribution priority determined, for described each virtual machine distributes physical resource.
12. as arbitrary in claim 9-11 as described in virtual machine manager, it is characterized in that, described allocation units, also for according to the distribution priority determined, before described each virtual machine distributes physical resource, determine each virtual machine that the type of service corresponding to service identification that reports is identical;
According to the priority corresponding to the business process mark in the traffic performance information that each virtual machine determined reports respectively, the distribution priority of the virtual machine that the priority corresponding to business process mark determining to report is high, higher than the distribution priority of the low virtual machine of the priority corresponding to the business process mark reported; Or
According to the Current resource utilization factor of each virtual machine determined, determine the distribution priority of the virtual machine that Current resource utilization factor is high, higher than the distribution priority of the low virtual machine of Current resource utilization factor.
13. as arbitrary in claim 9-11 as described in virtual machine manager, it is characterized in that, described allocation units, specifically for according to the Current resource utilization factor of described each virtual machine and the corresponding magnitude relationship setting resource utilization threshold, determine the physical resource amount of the current needs of described each virtual machine; And according to the physical resource amount of the current needs of each virtual machine determined, based on distribution priority order from high to low, be that described each virtual machine distributes physical resource.
14. as arbitrary in claim 9-11 as described in virtual machine manager, it is characterized in that, described allocation units, determine according to described traffic performance information the physical resource needed for described each virtual machine distributes specifically for working as, when exceeding the physical resource that described first server can distribute, in described each virtual machine, select part virtual machine and migrate on second server; And be that the virtual machine do not moved distributes physical resource based on the physical resource that described first server can be distributed; And distribute physical resource based on the virtual machine that the physical resource that described second server can distribute is migration.
15. virtual machine managers as claimed in claim 14, is characterized in that, described allocation units, specifically for according to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server; Or
The virtual machine that the physical resource selecting needs to distribute remains unchanged also migrates on second server; Or
According to the distribution priority determined, select distribute the low virtual machine of priority and migrate on second server in each virtual machine needing the physical resource distributed to remain unchanged.
16. 1 kinds of resource allocation systems, is characterized in that, comprising: virtualized each virtual machine and virtual machine manager on server, wherein:
Described virtual machine, for gathering the traffic performance information of the current process business of this virtual machine; And give described virtual machine manager by described traffic performance information reporting;
Described virtual machine manager, the described traffic performance information for reporting according to the described each virtual machine received is that described each virtual machine distributes physical resource;
Wherein, described virtual machine manager, specifically for the type of service corresponding to the service identification in the traffic performance information that reports respectively according to each virtual machine, determines being the distribution priority that described each virtual machine distributes when specifying physical resource; And according to the distribution priority determined, for described each virtual machine distributes described appointment physical resource.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010597688.8A CN102567072B (en) | 2010-12-20 | 2010-12-20 | Resource allocation method, resource allocation device and resource allocation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010597688.8A CN102567072B (en) | 2010-12-20 | 2010-12-20 | Resource allocation method, resource allocation device and resource allocation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102567072A CN102567072A (en) | 2012-07-11 |
| CN102567072B true CN102567072B (en) | 2015-04-22 |
Family
ID=46412571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010597688.8A Active CN102567072B (en) | 2010-12-20 | 2010-12-20 | Resource allocation method, resource allocation device and resource allocation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102567072B (en) |
Families Citing this family (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103037019B (en) * | 2013-01-07 | 2016-05-18 | 北京华胜天成科技股份有限公司 | A kind of distributed data acquisition system and method based on cloud computing |
| CN103220362A (en) * | 2013-04-23 | 2013-07-24 | 深圳市京华科讯科技有限公司 | Server virtualization all-in-one machine |
| CN104252390B (en) * | 2013-06-28 | 2018-08-14 | 华为技术有限公司 | Resource regulating method, device and system |
| US9870268B2 (en) * | 2013-08-05 | 2018-01-16 | Amazon Technologies, Inc. | Virtual computing instance migration |
| CN103455378A (en) * | 2013-09-05 | 2013-12-18 | 华为技术有限公司 | Resource distribution method of virtual clusters, device and system |
| CN103763346B (en) * | 2013-12-31 | 2017-07-25 | 华为技术有限公司 | A kind of distributed resource scheduling method and device |
| CN104866382B (en) * | 2014-02-21 | 2019-02-22 | 中国电信股份有限公司 | Schedule virtual resources method and apparatus |
| CN103795804A (en) | 2014-02-24 | 2014-05-14 | 华为技术有限公司 | Storage resource scheduling method and storage calculation system |
| CN105446813B (en) * | 2014-05-30 | 2018-12-14 | 华为技术有限公司 | A kind of method and device of resource allocation |
| CN106471771B (en) * | 2015-06-01 | 2019-06-21 | 华为技术有限公司 | A kind of method and network element device of congestion control |
| CN105162716A (en) * | 2015-07-28 | 2015-12-16 | 上海华为技术有限公司 | Flow control method and apparatus under NFV configuration |
| CN105591809B (en) * | 2015-10-12 | 2019-04-05 | 新华三技术有限公司 | A kind of resource allocation methods and device |
| CN105376111B (en) * | 2015-11-13 | 2019-04-26 | 百度在线网络技术（北京）有限公司 | Resource allocation methods and device |
| CN105471990B (en) * | 2015-11-23 | 2019-08-06 | 中国民生银行股份有限公司 | The construction method and device of characteristic service support platform |
| CN108702316B (en) * | 2016-03-30 | 2020-06-26 | 华为技术有限公司 | VNF resource allocation method and device |
| CN106020972A (en) * | 2016-05-10 | 2016-10-12 | 广东睿江云计算股份有限公司 | CPU (Central Processing Unit) scheduling method and device in cloud host system |
| CN105824708A (en) * | 2016-05-10 | 2016-08-03 | 广东睿江云计算股份有限公司 | CPU (Central Processing Unit) scheduling method and device in cloud host system |
| CN106020971B (en) * | 2016-05-10 | 2020-01-31 | 广东睿江云计算股份有限公司 | CPU scheduling method and device in cloud host system |
| CN105912408A (en) * | 2016-05-10 | 2016-08-31 | 广东睿江云计算股份有限公司 | Method and device for scheduling CPUs in cloud host systems |
| CN106020973A (en) * | 2016-05-10 | 2016-10-12 | 广东睿江云计算股份有限公司 | CPU (Central Processing Unit) scheduling method and device in cloud host system |
| CN106100872B (en) * | 2016-05-31 | 2019-04-05 | 中国联合网络通信集团有限公司 | A kind of business continuance support method and device |
| CN106126345A (en) * | 2016-06-30 | 2016-11-16 | 广东睿江云计算股份有限公司 | There is internal memory control method and the system of the physical machine of cloudy main frame |
| CN106445690A (en) * | 2016-10-17 | 2017-02-22 | 深圳市深信服电子科技有限公司 | Dynamic scheduling method and device for virtual machine resource |
| CN108123924B (en) * | 2016-11-30 | 2021-02-12 | 中兴通讯股份有限公司 | Resource management method and system |
| CN106789421B (en) * | 2016-12-17 | 2020-06-05 | 深圳中广核工程设计有限公司 | Collaborative design method and device |
| CN108965364B (en) * | 2017-05-22 | 2021-06-11 | 杭州海康威视数字技术股份有限公司 | Resource allocation method, device and system |
| CN107770236A (en) * | 2017-07-04 | 2018-03-06 | 中国电子科技集团公司电子科学研究院 | A kind of storage resource distribution method, device and computer-readable medium |
| CN107301088A (en) * | 2017-07-14 | 2017-10-27 | 郑州云海信息技术有限公司 | A kind of method and apparatus for managing virtual machine batch migration |
| CN107291557B (en) * | 2017-08-02 | 2020-10-23 | 中国银行股份有限公司 | Service processing method and device |
| CN107948095B (en) * | 2017-11-21 | 2021-11-02 | 中国银行股份有限公司 | Resource control method and device and bus system server |
| CN109960566A (en) * | 2017-12-26 | 2019-07-02 | 中国移动通信集团辽宁有限公司 | Distribution method, device, equipment and the medium of virtualization resource |
| CN109976870A (en) * | 2017-12-28 | 2019-07-05 | 中国移动通信集团重庆有限公司 | Creation method, device, equipment and the medium of virtual machine |
| CN108347391A (en) * | 2018-01-26 | 2018-07-31 | 全球能源互联网研究院有限公司 | A kind of network resource allocation method in network communication and device |
| CN109450684B (en) * | 2018-11-09 | 2020-06-30 | 北京邮电大学 | Method and device for expanding physical node capacity of network slicing system |
| CN110661654B (en) * | 2019-09-19 | 2023-02-28 | 北京浪潮数据技术有限公司 | Network bandwidth resource allocation method, device, equipment and readable storage medium |
| CN111258753A (en) * | 2020-01-08 | 2020-06-09 | 北京和信创天科技股份有限公司 | Method for guaranteeing priority allocation of virtual machine resources |
| CN112052075A (en) * | 2020-09-30 | 2020-12-08 | 中移（杭州）信息技术有限公司 | Cloud desktop virtual machine live migration method, server and storage medium |
| CN113032101B (en) * | 2021-03-31 | 2023-12-29 | 深信服科技股份有限公司 | Resource allocation method of virtual machine, server and computer readable storage medium |
| CN114416381B (en) * | 2022-03-28 | 2022-08-12 | 维塔科技(北京)有限公司 | Processing resource over-partitioning method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101488098A (en) * | 2009-02-13 | 2009-07-22 | 华中科技大学 | Multi-core computing resource management system based on virtual computing technology |
| CN101593133A (en) * | 2009-06-29 | 2009-12-02 | 北京航空航天大学 | Load balancing of resources of virtual machine method and device |
| CN101819540A (en) * | 2009-02-27 | 2010-09-01 | 国际商业机器公司 | Method and system for scheduling task in cluster |
| CN101876918A (en) * | 2009-11-27 | 2010-11-03 | 北京航空航天大学 | Synchronous task sensing method of virtual machine client operation system |
| CN101894047A (en) * | 2010-06-24 | 2010-11-24 | 北京航空航天大学 | Kernel virtual machine scheduling policy-based implementation method |
-
2010
- 2010-12-20 CN CN201010597688.8A patent/CN102567072B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101488098A (en) * | 2009-02-13 | 2009-07-22 | 华中科技大学 | Multi-core computing resource management system based on virtual computing technology |
| CN101819540A (en) * | 2009-02-27 | 2010-09-01 | 国际商业机器公司 | Method and system for scheduling task in cluster |
| CN101593133A (en) * | 2009-06-29 | 2009-12-02 | 北京航空航天大学 | Load balancing of resources of virtual machine method and device |
| CN101876918A (en) * | 2009-11-27 | 2010-11-03 | 北京航空航天大学 | Synchronous task sensing method of virtual machine client operation system |
| CN101894047A (en) * | 2010-06-24 | 2010-11-24 | 北京航空航天大学 | Kernel virtual machine scheduling policy-based implementation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102567072A (en) | 2012-07-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102567072B (en) | Resource allocation method, resource allocation device and resource allocation system | |
| Amini et al. | A Dynamic SLA Aware Heuristic Solution For IaaS Cloud Placement Problem Without Migration | |
| EP3507692B1 (en) | Resource oversubscription based on utilization patterns in computing systems | |
| US10334034B2 (en) | Virtual machine live migration method, virtual machine deployment method, server, and cluster system | |
| CN101593134B (en) | Method and device for allocating CPU resources of virtual machine | |
| KR101113943B1 (en) | Method And Apparatus For Managing Workload And Distributed Computing System Using The Same | |
| CN103051564B (en) | The method and apparatus of dynamic resource allocation | |
| CN112583861A (en) | Service deployment method, resource configuration method, system, device and server | |
| CN103473139A (en) | Virtual machine cluster resource allocation and scheduling method | |
| CN103856512A (en) | Cloud computing management server, working host, idle host and resource scheduling method | |
| CN109522090B (en) | Resource scheduling method and device | |
| CN110865881A (en) | Resource scheduling method and device | |
| CN105024842A (en) | Method and device for capacity expansion of server | |
| CN103179048A (en) | Method and system for changing main machine quality of service (QoS) strategies of cloud data center | |
| KR20110083084A (en) | Apparatus and method for operating server by using virtualization technology | |
| CN113608871A (en) | Service processing method and device | |
| KR20130019698A (en) | Method for optimizing resource by using migration based on user's scheduler | |
| US20120233313A1 (en) | Shared scaling server system | |
| CN110086726A (en) | A method of automatically switching Kubernetes host node | |
| CN105516267B (en) | Cloud platform efficient operation method | |
| CN109960579B (en) | Method and device for adjusting service container | |
| WO2013082742A1 (en) | Resource scheduling method, device and system | |
| CN103488538B (en) | Application extension device and application extension method in cloud computing system | |
| CN107203256A (en) | Energy-conservation distribution method and device under a kind of network function virtualization scene | |
| CN109144664B (en) | Dynamic migration method of virtual machine based on user service quality demand difference |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |