CN107045457A - A kind of method and device of multi-tenant resource-sharing - Google Patents
A kind of method and device of multi-tenant resource-sharing Download PDFInfo
- Publication number
- CN107045457A CN107045457A CN201610082020.7A CN201610082020A CN107045457A CN 107045457 A CN107045457 A CN 107045457A CN 201610082020 A CN201610082020 A CN 201610082020A CN 107045457 A CN107045457 A CN 107045457A
- Authority
- CN
- China
- Prior art keywords
- cpu core
- tenant
- cpu
- chained list
- core
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
Abstract
The present invention relates to field of computer technology, more particularly to a kind of method and device of multi-tenant resource-sharing, the problem of to solve current CPU core unreasonable distribution.This method includes：CPU management distributors estimate the average service rate of the CPU core of tenant, if it is determined that the average service rate of the CPU core of tenant is less than the expectation utilization rate of the CPU core pre-set for tenant, the corresponding CPU core information of at least one CPU core that then tenant is rented is added in memory space in default CPU core chained list, if it is determined that the average service rate of the CPU core of tenant is more than the expectation utilization rate of the CPU core pre-set, then distinguish corresponding CPU core at least one CPU core information that tenant is distributed in CPU core chained list, this technical scheme can dynamically distribute the CPU core that the tenant stored in memory space is rented by CPU management distributors, so as to the problem of avoiding current CPU core unreasonable distribution.
Description
Technical field
The present invention relates to field of computer technology, more particularly to a kind of method and device of multi-tenant resource-sharing.
Background technology
In cloud computing era, substantial amounts of computing resource is brought together, and uniformly provides the user service, user
Computing resource is rented from cloud, the benefit of low cost can have both been enjoyed, security, reliability can be obtained again
Guarantee, wherein, user is also known as tenant, a tenant generally one tissue of correspondence, such as company, learns
School etc..
Computing resource is with CPU (Central Processing Unit, central processing unit) core in the prior art
Divided for unit, all tenants obtain CPU core by competition mechanism, are so easily caused different rents
The CPU core that family is competed differs greatly, it is possible to which competition only needs less CPU to the more tenant of CPU core
Core can meet the demand of the tenant, and competing to the less tenant of CPU core needs more CPU core to expire
The demand of the foot tenant, causes CPU core resource allocation unreasonable, in addition, the ability of the competition CPU core of tenant
Difference over time, it may occur that change, such as network environment of the moment tenant preferably, are competed
To more CPU core, subsequent time network environment is deteriorated, for service the CPU core of the tenant be possible to for
Other tenants service, and cause CPU core switching times to increase, and the efficiency for handling tenant's demand is low.
The content of the invention
The present invention provides a kind of method and device of multi-tenant shared resource, to solve to exist in the prior art
CPU core unreasonable distribution and because tenant obtains CPU core, caused CPU by competition mechanism
Core switching times increase, the problem of processing tenant's demand efficiency is low.
First aspect there is provided a kind of method of multi-tenant shared resource, including：
Central processor CPU management distributor estimates the average service rate of the CPU core of tenant；If it is determined that described
The expectation that the average service rate of the CPU core of tenant is less than the CPU core pre-set for the tenant is used
The corresponding CPU core letter of at least one CPU core in rate, then each CPU core rented the tenant
Breath is added in the memory space in default CPU core chained list, the CPU stored in the CPU core chained list
The corresponding CPU core of nuclear information is used to redistribute；If it is determined that the average service rate of the CPU core of the tenant is big
In the expectation utilization rate of the CPU core pre-set, then distributed for the tenant in the CPU core chained list
At least one CPU core information distinguish corresponding CPU core, wherein, the average service rate of the CPU core
The average value for the utilization rate of each CPU core rented by the tenant in setting duration, the tenant is
Any tenant stored in the memory space of the CPU management distributor management.
The average service rate of the CPU core of tenant in setting duration can be estimated because CPU manages distributor, and
The expectation for being less than the CPU core pre-set for the tenant in the average service rate of the CPU core as the tenant makes
When using rate, the corresponding CPU core information addition of at least one CPU core in the CPU core that the tenant is rented
Into CPU core chained list, for redistributing, when it is determined that tenant needs to rent CPU core, then from CPU
The CPU core information that at least one has stored is selected in core chained list, and corresponding CPU core distributes to the tenant respectively,
So as to by the way that to tenant, the average service rate of CPU core is estimated in setting duration so that CPU management distribution
Device management can dynamically distribute the rent stored in memory space according to the average service rate of the tenant estimated
The CPU core that family is rented, thus avoid in the prior art caused by competition CPU core distribute it is unreasonable
The problem of, while the switching times of CPU core are reduced, so as to improve the efficiency of processing tenant's demand.
On the basis of first aspect, optionally, methods described also includes：The CPU management distributor exists
In the memory space during increased new tenant, then distributed for the new tenant in the CPU core chained list extremely
Few CPU core information distinguishes corresponding CPU core.
Due to that when increasing new tenant in memory space, can be distributed for the new group of family in CPU core chained list
At least one CPU core information distinguish corresponding CPU core, it is to avoid new user passes through competition mechanism, competition
CPU core, so as to avoid irrational problem that CPU core is distributed caused by competition in the prior art, together
When reduce the switching times of CPU core so that improve processing tenant's demand efficiency.
On the basis of first aspect, optionally, the CPU core chained list includes idle chained list and candidate's chained list；
When the average service rate of the CPU core of the tenant is less than the minimum of the CPU core pre-set for the tenant
During utilization rate, the CPU manages at least one in each CPU core that distributor is rented the tenant
The corresponding CPU core information of CPU core is added in the idle chained list；When the tenant CPU core it is flat
Equal utilization rate is more than the minimum utilization rate of the CPU core pre-set for the tenant and less than for described
During the expectation utilization rate for the CPU core that tenant pre-sets, the CPU management distributor is rented the tenant
The corresponding CPU core information of at least one CPU core in each CPU core, and the tenant rent
The rental precedence information mapping of CPU core is added in candidate's chained list；Wherein, the CPU of the tenant
The expectation utilization rate of core is more than the minimum utilization rate of the CPU core of the tenant.
Because the average service rate of the CPU core as the tenant estimated is less than the CPU pre-set for the tenant
During the minimum utilization rate of core, the tenant can reduce the number for renting CPU core, therefore will reduce the CPU rented
In the idle chained list of the nuclear information addition of core, and when the average service rate of the CPU core of the tenant estimated is more than pin
The minimum utilization rate of the CPU core pre-set to the tenant, less than the CPU core pre-set for the tenant
Expectation utilization rate when, illustrate that the tenant has demand to the CPU core of rental, can be in other tenants to CPU
When core has demand and can rented without idle CPU core, the CPU core of the tenant can be regard as one kind
Candidate scheme, by refining CPU core chained list, further such that the distribution for the CPU core rented tenant is more
Adduction is managed.
On the basis of first aspect, optionally, the CPU manages distributor if it is determined that the CPU of the tenant
The average service rate of core is more than the maximum utilization rate and the sky of the CPU core pre-set for the tenant
Not busy chained list is sky, then is selected according to the order of the rental precedence information stored in candidate's chained list from low to high
At least one selected out rents precedence information and distinguishes corresponding CPU core information, and the CPU core selected is believed
Cease corresponding CPU core and distribute to the tenant；If it is determined that the average service rate of the CPU core of the tenant is more than
The expectation utilization rate of the CPU core pre-set for the tenant and the idle chained list are not sky, then will
At least one the corresponding CPU core of CPU core information stored in the idle chained list distributes to the tenant；Its
In, the expectation utilization rate of the CPU core of the tenant is less than the maximum utilization rate of the CPU core of the tenant.
Because CPU manages distributor by the expectation utilization rate of the CPU core for the tenant and for the tenant
Maximum utilization rate, to distribute CPU core scheme further refined so that tenant is rented
The distribution of CPU core is more reasonable, and preferential be supplied to the low CPU core of tenant's priority needs CPU
Core tenant, so as to ensure that the use at high priority bank family.
On the basis of first aspect, optionally, the CPU manages distributor if it is determined that the CPU of the tenant
The average service rate of core is more than the expectation utilization rate of the CPU core pre-set for the tenant and less than pin
The maximum utilization rate of the CPU core pre-set to the tenant and the idle chained list are sky, then are no longer
The tenant distributes CPU core.
Because the program has carried out further refinement to the scheme for distributing CPU core so that tenant is rented
CPU core distribution it is more reasonable.
On the basis of first aspect, optionally, the CPU management distributors are if it is determined that it is at least two to need
Individual tenant distributes CPU core, then the rental priority of CPU core is rented according to each tenant at least two tenants
The order of information from high to low, respectively described each tenant's distribution CPU core.
It is preferably excellent because CPU management distributors are when it is determined that at least need for two tenant's distribution CPU cores
The high tenant's distribution CPU core of first level, therefore ensure that demand of the high tenant of priority for CPU core.
Second aspect there is provided a kind of device of multi-tenant shared resource, including：Computing module, judge mould
Block；Wherein, computing module is used for the average service rate for estimating the CPU core of tenant, and the CPU core is averaged
The average value of the utilization rate for each CPU core that utilization rate is rented by the tenant in setting duration, it is described
Tenant is that the CPU manages any tenant stored in the memory space of distributor management；Judge module is used
In it is determined that the average service rate of the CPU core of the tenant, which is less than, is directed to the CPU core that the tenant pre-sets
Expectation utilization rate when, then at least one CPU core correspondence in each CPU core rented the tenant
CPU core information be added in the memory space in default CPU core chained list, the CPU core chained list
The corresponding CPU core of CPU core information of middle storage is used to redistribute；It is determined that the CPU core of the tenant
Average service rate when being more than the expectation utilization rate of the CPU core pre-set, then be tenant distribution
At least one CPU core information in the CPU core chained list distinguishes corresponding CPU core.
Because computing module can estimate the average service rate of the CPU core of tenant in setting duration, and judge mould
Block is less than the expectation of the CPU core pre-set for the tenant in the average service rate of the CPU core as the tenant
During utilization rate, the corresponding CPU core information of at least one CPU core adds in the CPU core that the tenant is rented
It is added in CPU core chained list, for redistributing, when it is determined that tenant needs to rent CPU core, then from CPU
The CPU core information that at least one has stored is selected in core chained list, and corresponding CPU core distributes to the tenant respectively,
So as to by the way that to tenant, the average service rate of CPU core is estimated in setting duration so that the device being capable of root
According to the average service rate of the tenant estimated, dynamically distribute what the tenant stored in memory space was rented
CPU core, so as to avoid irrational problem that CPU core is distributed caused by competition in the prior art, together
When reduce the switching times of CPU core so that improve processing tenant's demand efficiency.
On the basis of second aspect, optionally, the device also includes：Processing module；The processing module
For during increased new tenant, then distributing the CPU core chained list for the new tenant in the memory space
In at least one CPU core information distinguish corresponding CPU core.
Due to that when increasing new tenant in memory space, can be distributed for the new group of family in CPU core chained list
At least one CPU core information distinguish corresponding CPU core, it is to avoid new user passes through competition mechanism, competition
CPU core, so as to avoid irrational problem that CPU core is distributed caused by competition in the prior art, together
When reduce the switching times of CPU core so that improve processing tenant's demand efficiency.
On the basis of second aspect, optionally, the CPU core chained list includes idle chained list and candidate's chained list；
The judge module, the average service rate specifically for the CPU core as the tenant is less than for the tenant
During the minimum utilization rate of the CPU core pre-set, in each CPU core that the tenant is rented at least
The corresponding CPU core information of one CPU core is added in the idle chained list；When the CPU core of the tenant
Average service rate be more than the minimum utilization rate of the CPU core pre-set for the tenant and less than being directed to
During the expectation utilization rate for the CPU core that the tenant pre-sets, each CPU core that the tenant is rented
In the corresponding CPU core information of at least one CPU core, and the tenant rent CPU core rental it is excellent
First level information MAP is added in candidate's chained list；Wherein, the expectation utilization rate of the CPU core of the tenant
More than the minimum utilization rate of the CPU core of the tenant.
Because the average service rate of the CPU core as the tenant estimated is less than the CPU pre-set for the tenant
During the minimum utilization rate of core, the tenant can reduce the number for renting CPU core, therefore will reduce the CPU rented
In the idle chained list of the nuclear information addition of core, and when the average service rate of the CPU core of the tenant estimated is more than pin
The minimum utilization rate of the CPU core pre-set to the tenant, less than the CPU core pre-set for the tenant
Expectation utilization rate when, illustrate that the tenant has demand to the CPU core of rental, can be in other tenants to CPU
When core has demand and can rented without idle CPU core, the CPU core of the tenant can be regard as one kind
Candidate scheme, by refining CPU core chained list, further such that the distribution for the CPU core rented tenant is more
Adduction is managed.
On the basis of second aspect, optionally, the judge module, specifically for if it is determined that the tenant
CPU core average service rate be more than the CPU core pre-set for the tenant maximum utilization rate and
The idle chained list is sky, then according to the rental precedence information stored in candidate's chained list from low to high
At least one rental precedence information that sequential selection goes out distinguishes corresponding CPU core information, by what is selected
The corresponding CPU core of CPU core information distributes to the tenant；If it is determined that the CPU core of the tenant is averaged
Utilization rate is more than the expectation utilization rate and the idle chained list of the CPU core pre-set for the tenant not
For sky, then at least one the corresponding CPU core of CPU core information stored in the idle chained list is distributed into institute
State tenant；Wherein, the expectation utilization rate of the CPU core of the tenant is less than the maximum of the CPU core of the tenant
Utilization rate.
Due to the expectation utilization rate by the CPU core for the tenant and the maximum utilization rate for the tenant,
The scheme for distributing CPU core is further refined so that the distribution for the CPU core rented tenant is more
Adduction is managed, and preferential be supplied to the low CPU core of tenant's priority needs CPU core tenant, so as to ensure
The use at high priority bank family.
On the basis of second aspect, optionally, the judge module is additionally operable to if it is determined that the tenant
The average service rate of CPU core is more than the expectation utilization rate of the CPU core pre-set for the tenant and small
In the CPU core pre-set for the tenant maximum utilization rate and the idle chained list for sky, then not
Again CPU core is distributed for the tenant.
Because the program has carried out further refinement to the scheme for distributing CPU core so that tenant is rented
CPU core distribution it is more reasonable.
On the basis of second aspect, optionally, the judge module is additionally operable to if it is determined that needing to be at least
Two tenants distribute CPU core, then the rental according to each tenant's rental CPU core at least two tenants is preferential
The order of level information from high to low, respectively described each tenant's distribution CPU core.
It is preferably excellent because CPU management distributors are when it is determined that at least need for two tenant's distribution CPU cores
The high tenant's distribution CPU core of first level, therefore ensure that demand of the high tenant of priority for CPU core.
Brief description of the drawings
Fig. 1 a are the system architecture schematic diagram of multi-tenant shared resource of the embodiment of the present invention；
Fig. 1 b are server architecture schematic diagram of the embodiment of the present invention；
Fig. 2 is the method flow schematic diagram of multi-tenant shared resource of the embodiment of the present invention；
Fig. 3 is the method flow schematic diagram of multi-tenant shared resource of the embodiment of the present invention；
Fig. 4 is the schematic diagram of the device of multi-tenant shared resource of the embodiment of the present invention；
Fig. 5 is the hardware architecture diagram of the device of multi-tenant shared resource of the embodiment of the present invention.
Embodiment
In order that the object, technical solutions and advantages of the present invention are clearer, with reference to Figure of description pair
The embodiment of the present invention is described in further detail.
The demand of the tenant is met to enable each tenant to be rented from high in the clouds to CPU core, is solved existing
The problem of CPU core unreasonable distribution present in technology and more CPU core switching times, the present invention is implemented
Example provides a kind of method of multi-tenant shared resource.
Fig. 1 a are the system architecture schematic diagram of multi-tenant shared resource of the embodiment of the present invention, and the system architecture includes：
Server and multiple terminal devices, wherein server as shown in Figure 1 b, including CPU management distributor and
Multiple CPU cores, wherein, tenant is by APP in terminal device (application, Application) to service
Device sends request, and it is that tenant distributes CPU core that server manages distributor by CPU, performs the request of tenant,
CPU manages the relevant information that distributor manages the memory space record tenant that distributor is managed by CPU,
The information includes but is not limited to the information of tenant itself, such as priority, title, and tenant uses CPU core
Usage record etc..
Using technical scheme provided in an embodiment of the present invention, due to CPU manages distributor can be according to estimating
The average service rate of tenant's CPU core, is dynamically managed by CPU and has been deposited in the memory space that distributor is managed
Storage tenant distribution CPU core so that avoid in the prior art caused by competition CPU core distribute do not conform to
The problem of reason, while the switching times of CPU core are reduced, so as to improve the efficiency of processing tenant's demand.
Technical scheme provided in an embodiment of the present invention is described in detail below.
As shown in Fig. 2 the method for multi-tenant shared resource of the embodiment of the present invention, including：
Step 200, CPU manages the average service rate that distributor estimates the CPU core of tenant, CPU core it is flat
The average value of the utilization rate for each CPU core that equal utilization rate is rented by tenant in setting duration, Zu Huwei
Any tenant stored in the memory space of CPU management distributor management；
Step 201, CPU manages distributor if it is determined that the average service rate of the CPU core of tenant is less than for renting
At least one in the expectation utilization rate for the CPU core that family is pre-set, then each CPU core rented tenant
The corresponding CPU core information of individual CPU core is added in memory space in default CPU core chained list, CPU core
The corresponding CPU core of CPU core information stored in chained list is used to redistribute；
Step 202, CPU manage distributor if it is determined that the CPU core of tenant average service rate be more than set in advance
The expectation utilization rate for the CPU core put, then at least one CPU core letter distributed for tenant in CPU core chained list
The corresponding CPU core of breath difference.
It should be noted that in step 200, a length of CPU management distributor estimates the first tenant during setting
CPU core average service rate after a period of time, i.e. CPU management distributor estimation tenant CPU
The average service rate of core is the average service rate of the CPU core of the tenant of following a period of time.
Specifically, CPU manages distributor by counting tenant's CPU core service data, to predict or estimate
The average service rate of CPU core of the tenant in setting duration.
Optionally, CPU manages the average service rate of the distributor periodically CPU core of the first tenant of estimation, tool
Body, CPU manages distributor when each cycle starts, and estimates the flat of the CPU core of the first tenant in the cycle
Equal utilization rate.
In step 201, when CPU management distributors determine that the average service rate of the CPU core of the tenant is less than
During the expectation utilization rate of the CPU core pre-set for the first tenant, according to default algorithm, first is determined
It can be used in the number of CPU core redistributed in the CPU core that tenant rents, and by the CPU of corresponding number
Nuclear information is added in CPU core chained list.
If CPU core chained list includes idle chained list and candidate's chained list, optionally, when the tenant CPU core it is flat
When equal utilization rate is less than the minimum utilization rate of the CPU core pre-set for the tenant, CPU management distributors
The corresponding CPU core information of at least one CPU core in each CPU core that the tenant is rented is added to
In idle chained list；When the average service rate of the CPU core of the tenant is more than the CPU pre-set for the tenant
When the minimum utilization rate of core and the expectation utilization rate less than the CPU core pre-set for the tenant, CPU
The corresponding CPU core of at least one CPU core in each CPU core that management distributor is rented the tenant
Information, and the rental precedence information mapping of tenant rental CPU core are added in candidate's chained list；Wherein,
The expectation utilization rate of the CPU core of the tenant is more than the minimum utilization rate of the CPU core of the tenant.
By taking tenant 1 as an example, the expectation utilization rate of the CPU core pre-set for tenant 1 is 70%, for
The minimum utilization rate for the CPU core that tenant 1 pre-sets is 50%：
Scene one：If the average service rate of the CPU core for the tenant 1 that CPU management distributors are estimated is 40%,
Because the average service rate of CPU core is less than the minimum utilization rate of the CPU core pre-set for tenant 1, then
According to the algorithm pre-set, it is determined that the number of the CPU core of release is 1, and by the CPU core of corresponding number
Nuclear information be added in idle chained list, specifically, determine release CPU core number be 1 the step of can
With as follows：
Tenant 1 needs the number for the CPU core rented as n=40%/50%*N in duration is set, and works as N=4
When, n=3.2 is rounded downwards, then the number for renting CPU core is needed in duration is set as 4, wherein, N
For the expectation number of the CPU core pre-set for tenant 1, if for the CPU core that current tenant 1 rents
Number is 5, then the number for the CPU core that can be discharged is 1.
Wherein, the corresponding CPU core of CPU core information preserved in idle chained list, refers to that tenant does not use
The CPU core, the CPU core is in idle condition.
It should be noted that determine that the algorithm of the number of the CPU core of release is not limited to above-mentioned algorithm, other energy
It is enough in the algorithm of the number for the CPU core for determining release.
Scene two：If the average service rate of the CPU core of the tenant of CPU management distributor estimations is 60%,
Because the average service rate of CPU core is more than minimum utilization rate and less than utilization rate is expected, then basis is set in advance
The algorithm put, is determined to the number of the CPU core for redistributing, and by the CPU core of corresponding number
The precedence information mapping that nuclear information and tenant rent CPU core is added in candidate's chained list.
Wherein, the corresponding CPU core of CPU core information stored in candidate's chained list, which is that tenant is current, also to be used
CPU core, as the tenant of application CPU core in need, can distribute to the CPU core needs to apply for CPU
The tenant of core.
In step 202, when it is determined that the average service rate of the CPU core of tenant is more than what is pre-set for tenant
During the expectation utilization rate of CPU core, further, according to default algorithm, determine that the tenant needs to apply renting
The number of CPU core.
Specifically, when the average service rate of the CPU core of tenant is more than the CPU core pre-set for the tenant
Expectation utilization rate and less than the maximum utilization rate of the CPU core pre-set for the tenant, and idle chain
Table is not space-time, then the CPU core for the determination number distributed for the tenant is store in idle chained list at least one
Individual CPU core information distinguishes corresponding CPU core, wherein, the maximum of the CPU core pre-set for tenant
Utilization rate is more than the expectation utilization rate of the CPU core pre-set for tenant.
When the expectation that the average service rate of the CPU core of tenant is more than the CPU core pre-set for the tenant makes
With rate and less than the maximum utilization rate of the CPU core pre-set for the tenant, and idle chained list is space-time,
Then no longer distribute CPU core for the second tenant.
When the average service rate of the CPU core of tenant is more than the most ambassador of the CPU core pre-set for the tenant
With rate, and idle chained list is not space-time, then the CPU core for the determination number distributed for the tenant is idle chained list
The CPU core information of middle storage distinguishes corresponding CPU core.
When the average service rate of the CPU core of tenant is more than the most ambassador of the CPU core pre-set for the tenant
With rate, and idle chained list is space-time, then during the CPU core of the determination number distributed for the tenant is candidate's chained list
The CPU core information of storage distinguishes corresponding CPU core.
It should be noted that working as CPU manages distributor if it is determined that needing to distribute CPU core at least two tenants,
Then the rental precedence information of CPU core from high to low suitable is rented according to each tenant at least two tenants
Sequence, respectively each tenant's distribution CPU core.
In addition, when increasing new tenant during CPU manages the memory space that distributor is managed, CPU management point
Orchestration distinguishes corresponding CPU core at least one CPU core information in new tenant's distribution CPU core chained list.
Optionally, when increasing new tenant during CPU manages the memory space that distributor is managed, CPU management
Distributor is then at least one CPU in the new idle chained list of tenant's distribution if it is determined that idle chained list is not sky
Nuclear information distinguishes corresponding CPU core；If it is determined that idle chained list is sky, then it is new tenant's distribution candidate's chained list
In at least one CPU core information distinguish corresponding CPU core, can so as to ensure that new tenant can be assigned to
The CPU core used.
Tenant's information is stored in the memory space that CPU management distributors are managed, as shown in figure 3, this hair
The method of bright embodiment multi-tenant shared resource, including：
Step 300, CPU manages the precedence information that distributor rents CPU core according to tenant, it is determined that storage
Tenant rents the order of CPU priority in space.
CPU management distributors determine the CPU core that can be used in redistributing in idle chained list and candidate's chained list
Information, CPU priority orders are rented according to tenant, the following steps are performed successively：
Step 301, CPU management distributor estimates being averaged for the CPU core of the tenant stored in memory space
Utilization rate, the average service rate of the CPU core is setting each CPU core rented in duration for the tenant
The average value of utilization rate.
Step 302, CPU management distributor judges whether the average service rate of the CPU core of the tenant is less than pin
The minimum utilization rate of the CPU core pre-set to the tenant, if so, then performing step 304, is otherwise performed
Step 303；
Step 303, CPU management distributor judges whether the average service rate of the CPU core of the tenant is less than pin
The expectation utilization rate of the CPU core pre-set to the tenant, if so, then performing step 305, is otherwise performed
Step 306；
Step 304, CPU manages the CPU core that distributor determines to discharge in the CPU core that the tenant rents
Number, and be possible to the CPU core information of corresponding number of release and be added in idle chained list, idle chained list
The corresponding CPU core of CPU core information of middle preservation is used to redistribute, and performs step 306.
Step 305, CPU manages distributor and determines to can be used in redistributing in the CPU core that the tenant rents
CPU core number, and be possible to CPU core information and the tenant of corresponding number for redistributing
The mapping for renting the precedence information of CPU core is added in candidate's chained list, the CPU core preserved in candidate's chained list
The corresponding CPU core of information is used to redistribute, and performs step 306.
Step 306, CPU management distributor judges whether memory space needs also exist for estimating in setting duration
The tenant of the average service rate of CPU core, if so, current tenant is switched to and current tenant's precedence information
Adjacent next tenant, re-executes step 301, otherwise performs step 307；
CPU manages the expectation that distributor is more than the CPU core pre-set according to the average service rate of CPU core
The priority orders of utilization rate tenant, are performed successively：
Step 307, CPU manages distributor it is determined that the average service rate of the CPU core of tenant is more than for being somebody's turn to do
During the expectation utilization rate that tenant pre-sets, determine that the tenant needs to apply for the number of the CPU core of rental；
Step 308, CPU manages distributor if it is determined that the average service rate of the CPU core of tenant is more than for being somebody's turn to do
The maximum utilization rate and free time chained list for the CPU core that tenant pre-sets not are sky, then are selected from idle chained list
The CPU core information for needing to apply for number is selected, and the CPU core information of selection is distinguished into corresponding CPU core point
The dispensing tenant uses.
Step 309, CPU manages distributor if it is determined that the average service rate of the CPU core of tenant is more than for being somebody's turn to do
The maximum utilization rate of the CPU core that tenant pre-sets and idle chained list is sky, then from candidate's chained list according to
Tenant rents CPU core priority from low to high, and selection needs to apply for the CPU core information of number, and will
The CPU core information of selection respectively distribute to the tenant and use by corresponding CPU core.
Step 310, CPU manages distributor if it is determined that the average service rate of the CPU core of tenant is more than for being somebody's turn to do
The expectation utilization rate for the CPU core that tenant pre-sets, less than the CPU core pre-set for the tenant most
Big utilization rate and idle chained list is sky, then no longer distribute CPU core for the tenant.
Step 311, CPU manages distributor if it is determined that the average service rate of the CPU core of tenant is more than for being somebody's turn to do
The expectation utilization rate for the CPU core that tenant pre-sets, less than the CPU core pre-set for the tenant most
Big utilization rate and idle chained list is not sky, then selection needs to apply for the CPU core letter of number from idle chained list
Breath, and corresponding CPU core is distributed to the tenant and used respectively by the CPU core information of selection.
Based on same inventive concept, a kind of device of multi-tenant shared resource is additionally provided in the embodiment of the present invention,
Because the corresponding method of device of multi-tenant shared resource of the embodiment of the present invention is a kind of multi-tenant shared resource
Method, therefore the implementation of the equipment of the embodiment of the present invention may refer to the implementation of method, repeats part and no longer goes to live in the household of one's in-laws on getting married
State.
As shown in figure 4, the device of multi-tenant shared resource of the embodiment of the present invention, including：
Computing module 400, the average service rate of the CPU core for estimating tenant, being averaged for CPU core is used
The average value of the utilization rate for each CPU core that rate is rented by tenant in setting duration, tenant manages for CPU
Any tenant stored in the memory space of reason distributor management；
Judge module 401, for it is determined that the average service rate of the CPU core of tenant is advance less than for tenant
At least one during the expectation utilization rate of the CPU core of setting, then in each CPU core rented tenant
The corresponding CPU core information of CPU core is added in memory space in default CPU core chained list, CPU core chain
The corresponding CPU core of CPU core information stored in table is used to redistribute；It is determined that the CPU core of tenant
When average service rate is more than the expectation utilization rate of the CPU core pre-set, then CPU core chained list is distributed for tenant
In at least one CPU core information distinguish corresponding CPU core.
The optional device also includes：Processing module 402, in memory space during increased new tenant,
Then distinguish corresponding CPU core at least one CPU core information that new tenant is distributed in CPU core chained list.
Optionally, CPU core chained list includes idle chained list and candidate's chained list；Judge module 401, specifically for
When the average service rate of the CPU core of tenant is less than the minimum utilization rate of the CPU core pre-set for tenant
When, the corresponding CPU core information addition of at least one CPU core in each CPU core that tenant is rented
Into idle chained list；When the average service rate of the CPU core of tenant is more than the CPU core pre-set for tenant
Minimum utilization rate and during the expectation utilization rate less than the CPU core pre-set for tenant, by tenant institute
The corresponding CPU core information of at least one CPU core in each CPU core rented, and tenant rent CPU
The rental precedence information mapping of core is added in candidate's chained list；Wherein, the expectation of the CPU core of tenant is used
Rate is more than the minimum utilization rate of the CPU core of tenant.
Optionally, judge module 401, specifically for if it is determined that the average service rate of the CPU core of tenant is more than
The maximum utilization rate of the CPU core pre-set for tenant and idle chained list is sky, then according to candidate's chained list
At least one rental precedence information point that the sequential selection of the rental precedence information of middle storage from low to high goes out
Not corresponding CPU core information, tenant is distributed to by the corresponding CPU core of CPU core information selected；If
Determine that expectation of the average service rate more than the CPU core pre-set for tenant of the CPU core of tenant is used
Rate and idle chained list is not sky, then by least one corresponding CPU core of CPU core information in idle chained list
Distribute to tenant；Wherein, the expectation utilization rate of the CPU core of tenant is used less than the maximum of the CPU core of tenant
Rate.
Optionally, judge module 401, are additionally operable to if it is determined that the average service rate of the CPU core of tenant is more than pin
The expectation utilization rate of the CPU core pre-set to tenant and less than the CPU core pre-set for tenant
Maximum utilization rate and idle chained list is sky, then no longer distribute CPU core for tenant.
Optionally, judge module 401, be additionally operable to if it is determined that need at least two tenants distribute CPU core,
Then the rental precedence information of CPU core from high to low suitable is rented according to each tenant at least two tenants
Sequence, respectively each tenant's distribution CPU core.
It should be noted that being schematical to the division of module in the embodiment of the present invention, only one kind is patrolled
Collect function to divide, there can be other dividing mode when actually realizing, in addition, in each embodiment of the application
In each functional module can be integrated in a processing module or individually be physically present, can also
Two or more modules are integrated in a module.Above-mentioned integrated module can both use the shape of hardware
Formula is realized, it would however also be possible to employ the form of software function module is realized.
If the integrated module realized using in the form of software function module and as independent production marketing or
In use, can be stored in a computer read/write memory medium.Understood based on such, the present invention
Part that the technical scheme of embodiment substantially contributes to prior art in other words or the technical scheme
It can completely or partially be embodied in the form of software product, the computer software product is stored in one and deposited
In storage media, including some instructions are to cause a computer equipment (can be personal computer, service
Device, or the network equipment etc.) or processor (processor) perform the application each embodiment methods described
All or part of step.And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disc
Or CD etc. is various can be with the medium of store program codes.
As shown in figure 5, the hardware architecture diagram of the device 50 for multi-tenant shared resource of the embodiment of the present invention,
Including processor 51, memory 52, input/output interface 53, communication interface 54 and bus 55, wherein,
Processor 51, memory 52, input/output interface 53 and communication interface 54 are realized each other by bus 55
Between communication connection, bus represents that the connected mode between other parts is only with thick line in Figure 5
Schematically illustrated, do not regard it as and be limited.The bus can be divided into address bus, data/address bus, control
Bus processed etc..For ease of representing, only represented in Fig. 5 with a thick line, it is not intended that only one bus
Or a type of bus.
Processor 51 can include general central processing unit (Central Processing Unit, CPU), micro-
Processor, application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or
The one or more integrated circuits of person, for performing relative program, to realize skill that the embodiment of the present invention is provided
Art scheme.
Memory 52 can be read-only storage (Read Only Memory, ROM), static storage device,
Dynamic memory or random access memory (Random Access Memory, RAM).Memory
52 can store an operating system, other applications and operational data.Realized by software or firmware
During technical scheme provided in an embodiment of the present invention, the journey for realizing technical scheme provided in an embodiment of the present invention
Sequence code is stored in memory 52, and is performed by processor 51.
From the above, it is seen that CPU management distributors estimate the average service rate of the CPU core of tenant,
The average service rate of CPU core is setting being averaged for the utilization rate for each CPU core rented in duration by tenant
Value, tenant is that CPU manages any tenant stored in the memory space of distributor management, however, it is determined that tenant
The average service rate of CPU core be less than the expectation utilization rate of the CPU core pre-set for tenant, then will rent
It is empty that the corresponding CPU core information of at least one CPU core in each CPU core that family is rented is added to storage
Between in default CPU core chained list, the corresponding CPU core of CPU core information stored in CPU core chained list is used
In redistributing, however, it is determined that the average service rate of the CPU core of tenant is more than the expectation of the CPU core pre-set
Utilization rate, then distinguish corresponding CPU at least one CPU core information that tenant is distributed in CPU core chained list
Core, this technical scheme due to CPU manage distributor management can according to the average service rate of the tenant estimated,
The CPU core that the tenant dynamically stored in distribution memory space is rented, so as to avoid in the prior art
Irrational problem that CPU core is distributed caused by competition, while the switching times of CPU core are reduced, from
And improve the efficiency of processing tenant's demand.
It should be understood by those skilled in the art that, embodiments of the invention can be provided as method, system or meter
Calculation machine program product.Therefore, the present invention can be using complete hardware embodiment, complete software embodiment or knot
The form of embodiment in terms of conjunction software and hardware.Wherein wrapped one or more moreover, the present invention can be used
Containing computer usable program code computer-usable storage medium (include but is not limited to magnetic disk storage,
CD-ROM, optical memory etc.) on the form of computer program product implemented.
The present invention is with reference to method according to embodiments of the present invention, equipment (system) and computer program product
Flow chart and/or block diagram describe.It should be understood that can by computer program instructions implementation process figure and/
Or each flow and/or square frame in block diagram and the flow in flow chart and/or block diagram and/or
The combination of square frame.These computer program instructions can be provided to all-purpose computer, special-purpose computer, embedded
Processor or the processor of other programmable data processing devices are to produce a machine so that pass through computer
Or the instruction of the computing device of other programmable data processing devices is produced for realizing in one stream of flow chart
The device for the function of being specified in journey or multiple flows and/or one square frame of block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices
In the computer-readable memory worked in a specific way so that be stored in the computer-readable memory
Instruction, which is produced, includes the manufacture of command device, and the command device is realized in one flow of flow chart or multiple streams
The function of being specified in one square frame of journey and/or block diagram or multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices, made
Obtain and perform series of operation steps on computer or other programmable devices to produce computer implemented place
Reason, so that the instruction performed on computer or other programmable devices is provided for realizing in flow chart one
The step of function of being specified in flow or multiple flows and/or one square frame of block diagram or multiple square frames.
, but those skilled in the art once know base although preferred embodiments of the present invention have been described
This creative concept, then can make other change and modification to these embodiments.So, appended right will
Ask and be intended to be construed to include preferred embodiment and fall into having altered and changing for the scope of the invention.
Obviously, those skilled in the art can carry out various changes and modification without departing from this hair to the present invention
Bright spirit and scope.So, if the present invention these modifications and variations belong to the claims in the present invention and
Within the scope of its equivalent technologies, then the present invention is also intended to comprising including these changes and modification.
Claims (12)
1. a kind of method of multi-tenant shared resource, it is characterised in that including：
Central processor CPU management distributor estimates the average service rate of the CPU core of tenant, the CPU
The average service rate of core is setting being averaged for the utilization rate for each CPU core rented in duration by the tenant
Value, the tenant is that the CPU manages any tenant stored in the memory space of distributor management；
The CPU management distributors are if it is determined that the average service rate of the CPU core of the tenant is less than for described
In the expectation utilization rate for the CPU core that tenant pre-sets, then each CPU core rented the tenant
The corresponding CPU core information of at least one CPU core is added to default CPU core chained list in the memory space
In, the corresponding CPU core of CPU core information stored in the CPU core chained list is used to redistribute；
The CPU management distributors are if it is determined that the average service rate of the CPU core of the tenant is advance more than described
The expectation utilization rate of the CPU core of setting, then distribute at least one in the CPU core chained list for the tenant
CPU core information distinguishes corresponding CPU core.
2. the method as described in claim 1, it is characterised in that methods described also includes：
The CPU management distributor is then the new tenant during increased new tenant in the memory space
At least one the CPU core information distributed in the CPU core chained list distinguishes corresponding CPU core.
3. method as claimed in claim 1 or 2, it is characterised in that the CPU core chained list includes sky
Not busy chained list and candidate's chained list；
At least one CPU in each CPU core that the CPU management distributor is rented the tenant
The corresponding CPU core information of core is added in CPU core chained list, including：
When the average service rate of the CPU core of the tenant is less than the CPU core pre-set for the tenant
During minimum utilization rate, in each CPU core that the CPU management distributor is rented the tenant at least
The corresponding CPU core information of one CPU core is added in the idle chained list；
When the average service rate of the CPU core of the tenant is more than the CPU core pre-set for the tenant
When minimum utilization rate and the expectation utilization rate less than the CPU core pre-set for the tenant, the CPU
The corresponding CPU of at least one CPU core in each CPU core that management distributor is rented the tenant
Nuclear information, and the rental precedence information mapping of tenant rental CPU core are added to candidate's chained list
In；
Wherein, the expectation utilization rate of the CPU core of the tenant is used more than the minimum of the CPU core of the tenant
Rate.
4. method as claimed in claim 3, it is characterised in that the CPU management distributor is described
At least one CPU core information that tenant is distributed in the CPU core chained list distinguishes corresponding CPU core, including：
The CPU management distributors are if it is determined that the average service rate of the CPU core of the tenant is more than for described
The maximum utilization rate for the CPU core that tenant pre-sets and the idle chained list are sky, then according to the candidate
At least one rental priority letter that the sequential selection of the rental precedence information stored in chained list from low to high goes out
The corresponding CPU core information of breath difference, the corresponding CPU core of CPU core information selected is distributed to described
Tenant；
The CPU management distributors are if it is determined that the average service rate of the CPU core of the tenant is more than for described
The expectation utilization rate for the CPU core that tenant pre-sets and the idle chained list are not sky, then by the free time
At least one corresponding CPU core of CPU core information in chained list distributes to the tenant；
Wherein, the expectation utilization rate of the CPU core of the tenant is used less than the maximum of the CPU core of the tenant
Rate.
5. the method as described in claim 3 or 4, it is characterised in that methods described also includes：
The CPU management distributors are if it is determined that the average service rate of the CPU core of the tenant is more than for described
The expectation utilization rate for the CPU core that tenant pre-sets and less than the CPU core pre-set for the tenant
Maximum utilization rate and the idle chained list for sky, then no longer for the tenant distribute CPU core.
6. the method as described in claim 1 to 5 is any, it is characterised in that the CPU management distribution
Device is that the tenant distributes CPU core, including：
The CPU management distributors are if it is determined that need to distribute CPU core at least two tenants, then according at least
Each tenant rents the order of the rental precedence information of CPU core from high to low, respectively institute in two tenants
State each tenant's distribution CPU core.
7. a kind of device of multi-tenant shared resource, it is characterised in that including：
Computing module, the average service rate of the CPU core for estimating tenant, being averaged for the CPU core is used
The average value of the utilization rate for each CPU core that rate is rented by the tenant in setting duration, the tenant
Any tenant stored in the memory space of distributor management is managed for the CPU；
Judge module, for it is determined that the average service rate of the CPU core of the tenant, which is less than, is directed to the tenant
During the expectation utilization rate of the CPU core pre-set, then in each CPU core rented the tenant extremely
Few corresponding CPU core information of a CPU core is added in the memory space in default CPU core chained list,
The corresponding CPU core of CPU core information stored in the CPU core chained list is used to redistribute；
The judge module, is additionally operable to it is determined that the average service rate of the CPU core of the tenant is more than described pre-
During the expectation utilization rate of the CPU core first set, then distributed for the tenant in the CPU core chained list at least
One CPU core information distinguishes corresponding CPU core.
8. device as claimed in claim 7, it is characterised in that also include：Processing module；
The processing module, for being then the new tenant during increased new tenant in the memory space
At least one the CPU core information distributed in the CPU core chained list distinguishes corresponding CPU core.
9. device as claimed in claim 7 or 8, it is characterised in that the CPU core chained list includes sky
Not busy chained list and candidate's chained list；
The judge module, specifically for：
When the average service rate of the CPU core of the tenant is less than the CPU core pre-set for the tenant
During minimum utilization rate, at least one CPU core in each CPU core that the tenant is rented is corresponding
CPU core information is added in the idle chained list；When the average service rate of the CPU core of the tenant is more than pin
The minimum utilization rate of the CPU core pre-set to the tenant and pre-set less than for the tenant
During the expectation utilization rate of CPU core, at least one CPU in each CPU core that the tenant is rented
The corresponding CPU core information of core, and the rental precedence information mapping of tenant rental CPU core are added to
In candidate's chained list；Wherein, the expectation utilization rate of the CPU core of the tenant is more than the CPU of the tenant
The minimum utilization rate of core.
10. device as claimed in claim 9, it is characterised in that the judge module, specifically for：
If it is determined that the average service rate of the CPU core of the tenant is more than the CPU pre-set for the tenant
The maximum utilization rate of core and the idle chained list are sky, then excellent according to the rental stored in candidate's chained list
At least one rental precedence information that the sequential selection of first level information from low to high goes out distinguishes corresponding CPU core
Information, the tenant is distributed to by the corresponding CPU core of CPU core information selected；If it is determined that the tenant
CPU core average service rate be more than the CPU core pre-set for the tenant expectation utilization rate and
The idle chained list is not sky, then by the corresponding CPU of at least one CPU core information in the idle chained list
Core distributes to the tenant；Wherein, the expectation utilization rate of the CPU core of the tenant is less than the CPU of the tenant
The maximum utilization rate of core.
11. the device as described in claim 9 or 10, it is characterised in that the judge module, is additionally operable to：
If it is determined that the average service rate of the CPU core of the tenant is more than the CPU pre-set for the tenant
The expectation utilization rate of core and the maximum utilization rate less than the CPU core pre-set for the tenant and institute
It is sky to state idle chained list, then no longer distributes CPU core for the tenant.
12. the device as described in claim 7 to 11 is any, it is characterised in that the judge module, also
For：
If it is determined that needing to distribute CPU core at least two tenants, then according to each tenant at least two tenants
The order of the rental precedence information of CPU core from high to low is rented, respectively described each tenant's distribution CPU
Core.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610082020.7A CN107045457A (en) | 2016-02-05 | 2016-02-05 | A kind of method and device of multi-tenant resource-sharing |
| PCT/CN2017/071022 WO2017133421A1 (en) | 2016-02-05 | 2017-01-13 | Method and device for sharing resources among multiple tenants |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610082020.7A CN107045457A (en) | 2016-02-05 | 2016-02-05 | A kind of method and device of multi-tenant resource-sharing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107045457A true CN107045457A (en) | 2017-08-15 |
Family
ID=59499290
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610082020.7A Withdrawn CN107045457A (en) | 2016-02-05 | 2016-02-05 | A kind of method and device of multi-tenant resource-sharing |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107045457A (en) |
| WO (1) | WO2017133421A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110780992A (en) * | 2019-09-20 | 2020-02-11 | 苏州浪潮智能科技有限公司 | Cloud computing platform optimized deployment method, system, terminal and storage medium |
| CN111597034A (en) * | 2019-02-21 | 2020-08-28 | 阿里巴巴集团控股有限公司 | Processor resource scheduling method and device, terminal equipment and computer storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100990412B1 (en) * | 2009-10-29 | 2010-10-29 | 주식회사 팀스톤 | Computer server capable of supporting cpu virtualization |
| CN102317917B (en) * | 2011-06-30 | 2013-09-11 | 华为技术有限公司 | Hot field virtual machine cpu dispatching method and virtual machine system (vms) |
| CN102946433B (en) * | 2012-11-22 | 2015-07-29 | 合肥华云通信技术有限公司 | The monitoring of the large-scale computer resource under cloud public service platform and dispatching method |
| US9602426B2 (en) * | 2013-06-21 | 2017-03-21 | Microsoft Technology Licensing, Llc | Dynamic allocation of resources while considering resource reservations |
| CN104111876A (en) * | 2014-07-10 | 2014-10-22 | 深圳供电局有限公司 | Dynamic resource management device and method based on Oracle resource plan |
-
2016
- 2016-02-05 CN CN201610082020.7A patent/CN107045457A/en not_active Withdrawn
-
2017
- 2017-01-13 WO PCT/CN2017/071022 patent/WO2017133421A1/en active Application Filing
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111597034A (en) * | 2019-02-21 | 2020-08-28 | 阿里巴巴集团控股有限公司 | Processor resource scheduling method and device, terminal equipment and computer storage medium |
| CN111597034B (en) * | 2019-02-21 | 2023-04-28 | 阿里巴巴集团控股有限公司 | Processor resource scheduling method and device, terminal equipment and computer storage medium |
| CN110780992A (en) * | 2019-09-20 | 2020-02-11 | 苏州浪潮智能科技有限公司 | Cloud computing platform optimized deployment method, system, terminal and storage medium |
| CN110780992B (en) * | 2019-09-20 | 2023-01-10 | 苏州浪潮智能科技有限公司 | Cloud computing platform optimized deployment method, system, terminal and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2017133421A1 (en) | 2017-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104102543B (en) | The method and apparatus of adjustment of load in a kind of cloud computing environment | |
| CN105900064B (en) | The method and apparatus for dispatching data flow task | |
| CN109343965A (en) | Resource adjusting method, device, cloud platform and server | |
| CN107239336B (en) | Method and device for realizing task scheduling | |
| US9277002B2 (en) | Physical resource management | |
| CN112153700A (en) | Network slice resource management method and equipment | |
| CN104572301B (en) | A kind of resource allocation methods and system | |
| WO2017166643A1 (en) | Method and device for quantifying task resources | |
| CN108268317A (en) | A kind of resource allocation methods and device | |
| CN102970379A (en) | Method for realizing load balance among multiple servers | |
| CN106407207A (en) | Real-time added data updating method and apparatus | |
| CN104461673B (en) | A kind of virtual machine (vm) migration determination method and device | |
| CN109491788A (en) | A kind of virtual platform implementation of load balancing and device | |
| CN109522090A (en) | Resource regulating method and device | |
| CN103593224A (en) | Virtual machine resource allocation system and method | |
| CN106775470B (en) | Data storage method and system | |
| CN108376103A (en) | A kind of the equilibrium of stock control method and server of cloud platform | |
| CN110275760A (en) | Process based on fictitious host computer processor hangs up method and its relevant device | |
| CN108241535B (en) | Resource management method and device and server equipment | |
| CN107045457A (en) | A kind of method and device of multi-tenant resource-sharing | |
| CN116302453B (en) | Task scheduling method and device for quantum electronic hybrid platform | |
| CN106020971B (en) | CPU scheduling method and device in cloud host system | |
| CN107402812A (en) | Cluster resource dispatching method, device, equipment and storage medium | |
| CN110286961A (en) | Process based on physical host processor hangs up method and relevant device | |
| CN108429704B (en) | Node resource allocation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170815 |
|
| WW01 | Invention patent application withdrawn after publication |