CN109213588A - A kind of cloud data center Batch Arrival task allocation apparatus, system and method - Google Patents
A kind of cloud data center Batch Arrival task allocation apparatus, system and method Download PDFInfo
- Publication number
- CN109213588A CN109213588A CN201811079294.6A CN201811079294A CN109213588A CN 109213588 A CN109213588 A CN 109213588A CN 201811079294 A CN201811079294 A CN 201811079294A CN 109213588 A CN109213588 A CN 109213588A
- Authority
- CN
- China
- Prior art keywords
- host
- task
- data center
- load
- cloud data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- 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
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
-
- 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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
Abstract
The invention discloses a kind of cloud data center Batch Arrival task allocation apparatus, system and method, it is related to cloud computing system control field, the present invention is assessed by the demand in real time to new arrival task to computing resource, and according to cloud data center currently each host operation load, reasonable distribution method is made to batch tasks, to reduce Task Congestion appearance, load balancing, the operational efficiency of lifting system entirety are realized.
Description
Technical field
The invention belongs to cloud computing system control fields, distribute more particularly to a kind of cloud data center Batch Arrival task
Device, system and method.
Background technique
Cloud computing is a kind of calculation Internet-based, in this way, shared software and hardware resources and information
It can be supplied to computer and other equipment on demand.Relative to traditional software and form is calculated, cloud computing has loose coupling
The significant advantages such as conjunction, on-demand, cost is controllable, resource is virtual, isomery collaboration, make its more adapt to e-commerce now,
The application such as flexible manufacturing, mobile Internet.Cloud data center refer to it is by multiple isomeries, by being connected to the network host institute together
The distributed computing system for being used to carry the enterprise-level application that online cloud service is provided of composition.It, will be big in cloud data center
The host of amount carries out centralized and unified management, can ensure that host runs required stabilized power supply environment, suitable temperature and humidity control
System and network bandwidth conditions.
The same with other software and hardware systems, the load of the host in cloud data center is also in real-time change.Due to existing
Modern cloud computing system applies to high loads and the height such as extensive scientific algorithm, real time financial, online transaction, Streaming Media multicast more
The application of complexity, host are in the state of overload operation often.If task schedule and load balancing are slightly not
It is careful, just easily there is the case where part host node waiting task excess accumulation and buffer overflow, and be possible to further
System crash is formed, heavy losses are caused.And traditional task schedule and allocation strategy, mainly only consider the money of alternative host
The match condition of source surplus and mission requirements amount carries out task distribution, and according to these static datas determine task distribution and
Scheduling scheme.This strategy has the disadvantage that due to ignoring load on host computers, resources occupation rate, availability and reliability
A large amount of newly-increased task may be assigned to that current task surplus is larger but reliability deteriorates by variation tendency, traditional scheduling strategy
In host on, easily make these hosts formed over time load hot spot；Meanwhile although more than the host current task of part
Amount is little, but during rapid improvement in operating status, and traditional dispatching distribution strategy is possible to neglect these
The host of more multitask can actually be accepted, so as to cause the idle waste of system resources in computation.
Summary of the invention
In view of the above drawbacks of the prior art, technical problem to be solved by the invention is to provide in a kind of pair of cloud data
The method of heart Batch Arrival task progress reasonable distribution.
To achieve the above object, the present invention provides a kind of cloud data center Batch Arrival method for allocating tasks, features
It is to sequentially include the following steps:
Step 1: monitoring each host current state of cloud data center:
The host number in cloud data center is set as n, 0 < i≤n, n > 0；
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement to task centering centre processing quantity
XQCPUj, each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud
Each available central processing unit quantity SYCPU of host current residual in data centeri, each host is current in cloud data center
Remaining free memory capacity SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, on each host
Time RWXH consumed by one step completed taski, each host last time complete task to current interval time WCJGi, it is each
A host the last time trouble duration GZCXi, each host the last time failure to current interval time GZJGi, it is each
The host last time completes the central processing unit quantity SFCPU that task is dischargedi, each host last time complete what task was discharged
Memory size SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Step 2: assessing each load on host computers:
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive execution of task of each host
Rate；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of a hosti；
A4, equivalent expected future implementation rate is set as DXZXi:
It calculatesIt obtains
Imitate expected future implementation rate；
It is assessed Step 3: accepting task priority to each host:
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；
A is time-concerning impact factor；
Step 4: generating batch tasks allocation plan:
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYWCx≥XQWCiObtain first
Collection is combined Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
Step 5: executing the distribution of batch weight task:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the use of submission task
Family；When by the corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribution
To MBjNumber host executes.
Preferably, being calculated in the step 2
Obtain the abundant resource degree of each host ladder
CPUCYi、WCCYi、NCCYi。
Preferably, in the step 4, the PDijThe two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
Using above technical scheme, by considering the dynamic fluctuation of cloud data center operating status, each host is worked as
Preceding task margin and the real-time situation of change of task load further improve the conjunction of cloud data center Batch Arrival task distribution
Rationality.
It is carried out rationally another technical problem to be solved by the present invention is that providing a kind of pair of cloud data center Batch Arrival task
Assigned unit.
To achieve the goals above, the present invention provides a kind of cloud data center Batch Arrival task allocation apparatus, including
Load on host computers monitoring modular, " task-host " matching module and batch tasks distribute execution module；The load on host computers monitors mould
Block includes load on host computers acquiring unit；" task-host " matching module is commented by load on host computers assessment unit, chiller priority degree
Estimate unit, batch tasks allocation strategy generation unit composition；Described in the first output end connection of the load on host computers acquiring unit
The input terminal of load on host computers assessment unit, the second output terminal of the load on host computers acquiring unit connect the chiller priority degree and comment
Estimate unit first input end, the third output end of the load on host computers acquiring unit connects the batch tasks allocation strategy and generates
Unit first input end, the output end of the load on host computers assessment unit connect the chiller priority degree assessment unit second and input
End, the chiller priority degree assessment unit output end connect the second input terminal of the batch tasks allocation strategy generation unit,
The output end of the batch tasks allocation strategy generation unit connects the batch tasks and distributes execution module input terminal；
The load on host computers acquiring unit is for obtaining each host current state of data center, and it is negative to be sent to the host
Carry assessment unit, chiller priority degree assessment unit, batch tasks allocation strategy generation unit:
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement to task centering centre processing quantity
XQCPUj, each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud
Each available central processing unit quantity SYCPU of host current residual in data centeri, each host is current in cloud data center
Remaining free memory capacity SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, on each host
Time RWXH consumed by one step completed taski, each host last time complete task to current interval time WCJGi, it is each
A host the last time trouble duration GZCXi, each host the last time failure to current interval time GZJGi, it is each
The host last time completes the central processing unit quantity SFCPU that task is dischargedi, each host last time complete what task was discharged
Memory size SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Calculated variate-value is sent to the master for assessing each load on host computers by the load on host computers assessment unit
Machine priority assessment unit:
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive execution of task of each host
Rate；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of a hosti；
A4, equivalent expected future implementation rate is set as DXZXi:
It calculatesIt obtains
Imitate expected future implementation rate；
The chiller priority degree assessment unit accepts task priority for assessing each host, and by calculated variate-value
It is sent to the batch tasks allocation strategy generation unit:
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；a
For time-concerning impact factor；
The batch tasks allocation strategy generation unit is for generating batch tasks allocation plan, and by calculated variable
Value is sent to the batch tasks distribution execution module:
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYWCx≥XQWCiObtain first
Collection is combined Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
The batch tasks distribution execution module is for executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the use of submission task
Family；When by the corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribution
To MBjNumber host executes.
Preferably, calculating
Obtain the abundant resource degree of each host ladder
CPUCYi、WCCYi、NCCYi。
Preferably, the PDijThe two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
The present invention also technical problems to be solved are to provide a kind of pair of cloud data center Batch Arrival task and are rationally divided
Match system
To achieve the above object, the present invention provides a kind of cloud data center Batch Arrival task distribution systems, including cloud
The server of data center is provided with cloud data center Batch Arrival task allocation apparatus, the cloud data in the server
Center Batch Arrival task allocation apparatus includes load on host computers monitoring modular, " task-host " matching module and batch tasks point
With execution module；The load on host computers monitoring modular includes load on host computers acquiring unit；" task-host " matching module by
Load on host computers assessment unit, chiller priority degree assessment unit, batch tasks allocation strategy generation unit composition；The load on host computers
First output end of acquiring unit connects the input terminal of the load on host computers assessment unit, and the of the load on host computers acquiring unit
Two output ends connect the chiller priority degree assessment unit first input end, the third output end of the load on host computers acquiring unit
The batch tasks allocation strategy generation unit first input end is connected, the output end of the load on host computers assessment unit connects institute
Chiller priority degree the second input terminal of assessment unit is stated, the chiller priority degree assessment unit output end connects the batch tasks point
The second input terminal with strategy generating unit, the output end of the batch tasks allocation strategy generation unit connect the batch and appoint
Business distribution execution module input terminal；
The load on host computers acquiring unit is for obtaining each host current state of data center, and it is negative to be sent to the host
Carry assessment unit, chiller priority degree assessment unit, batch tasks allocation strategy generation unit:
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement to task centering centre processing quantity
XQCPUj, each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud
Each available central processing unit quantity SYCPU of host current residual in data centeri, each host is current in cloud data center
Remaining free memory capacity SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, on each host
Time RWXH consumed by one step completed taski, each host last time complete task to current interval time WCJGi, it is each
A host the last time trouble duration GZCXi, each host the last time failure to current interval time GZJGi, it is each
The host last time completes the central processing unit quantity SFCPU that task is dischargedi, each host last time complete what task was discharged
Memory size SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Calculated variate-value is sent to by the load on host computers assessment unit (201) for assessing each load on host computers
The chiller priority degree assessment unit (202):
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive execution of task of each host
Rate；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of a hosti；
A4, equivalent expected future implementation rate is set as DXZXi:
It calculatesIt obtains
Imitate expected future implementation rate；
The chiller priority degree assessment unit (202) accepts task priority for assessing each host, and will be calculated
Variate-value is sent to the batch tasks allocation strategy generation unit (203):
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；
A is time-concerning impact factor；
The batch tasks allocation strategy generation unit (203) will calculate for generating batch tasks allocation plan
Variate-value be sent to batch tasks distribution execution module (3):
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYWCx≥XQWCiObtain first
Collection is combined Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
Batch tasks distribution execution module (3) is for executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the use of submission task
Family；When by the corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribution
To MBjNumber host executes.
Preferably, calculating
Obtain the abundant resource degree of each host ladder
CPUCYi、WCCYi、NCCYi。
Preferably, the PDijThe two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
The beneficial effects of the present invention are: the present invention has fully considered the dynamic fluctuation of cloud data center operating status, needle
To the real-time situation of change of current task surplus and task load of each host, Batch Arrival task is reasonably commented
Estimate, sort and task distribution, realize dynamic load balancing, avoids Task Congestion and load hot spot being emerged.
Detailed description of the invention
Fig. 1 is the flow diagram of the embodiment of the invention.
Fig. 2 is the structural schematic block diagram of another specific embodiment of the present invention.
Fig. 3 is the structural schematic block diagram of another specific embodiment of the present invention.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples:
As shown in Figure 1, a kind of cloud data center Batch Arrival method for allocating tasks, sequentially includes the following steps:
Step 1: monitoring each host current state of cloud data center:
The host number in cloud data center is set as n, 0 < i≤n, n > 0；
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement to task centering centre processing quantity
XQCPUj, each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud
Each available central processing unit quantity SYCPU of host current residual in data centeri, each host is current in cloud data center
Remaining free memory capacity SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, on each host
Time RWXH consumed by one step completed taski, each host last time complete task to current interval time WCJGi, it is each
A host the last time trouble duration GZCXi, each host the last time failure to current interval time GZJGi, it is each
The host last time completes the central processing unit quantity SFCPU that task is dischargedi, each host last time complete what task was discharged
Memory size SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Step 2: assessing each load on host computers:
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
Above scheme, the total abundant intensity of the resource of i-th of host, by remaining processor, remaining external memory, free memory ladder
The abundant resource degree of change codetermines.As long as if there is a kind of residue of resource in remaining processor, remaining external memory, free memory
Amount is less than the average residual amount of n host, then the abundant resource degree of ladder is calculated as 0.
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive execution of task of each host
Rate；
Above scheme, time RWXH consumed by the task that each host last time completesiCentainly it is not 0, therefore does not deposit
The case where divisor is zero.
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of a hosti；
Above scheme considers the Host Idle time after the completion of last task, free time more long then systemic energy loss
The reparation of consumption is more abundant, and the task execution rate of expected future is higher.
A4, the equivalent expected future implementation rate for considering fault effect is set as DXZXi:
It calculatesIt is examined
Consider the equivalent expected future implementation rate of fault effect；
It is assessed Step 3: accepting task priority to each host:
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；a
For time-concerning impact factor；A is 0.5；
The priority of above scheme, host is determined by the product of remaining computing resource abundant intensity and equivalent expected future implementation rate
It is fixed, while being completed interval time and nearest time between failures by nearest task and influenced, and recently time between failures with
Priority is negatively correlated, and nearest task completes interval time and priority is positively correlated.
Step 4: generating batch tasks allocation plan:
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYWCx≥XQWCiObtain first
Collection is combined Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj: For empty set；As calculated MBjWhen not being 0,It is rewritten as 1；
Above scheme, if the surplus resources of none host are able to satisfy the demand of task, corresponding target master
Machine labelled amount is calculated as 0；If there is the surplus resources of at least one host meet the needs of task, then there will be maximum priority
Host be considered as destination host.Meanwhile above-mentioned mechanism guarantees that different tasks is not assigned on the same host, to ensure
Load balancing.
Step 5: executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the use of submission task
Family；When by the corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribution
To MBjNumber host executes.
Then, it returns again to and executes step 1 detection each host current state of cloud data center.
In the present embodiment, calculate
Obtain the abundant resource degree of each host ladder
CPUCYi、WCCYi、NCCYi。
In the present embodiment, the PDijThe two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
As shown in Fig. 2, a kind of cloud data center Batch Arrival task allocation apparatus, including load on host computers monitoring modular 1,
" task-host " matching module 2 and batch tasks distribute execution module 3；The load on host computers monitoring modular 1 includes load on host computers
Acquiring unit 101；" task-host " matching module 2 is by load on host computers assessment unit 201, chiller priority degree assessment unit
202, batch tasks allocation strategy generation unit 203 forms；First output end of the load on host computers acquiring unit 101 connects institute
The input terminal of load on host computers assessment unit 201 is stated, the second output terminal of the load on host computers acquiring unit 101 connects the host
202 first input end of priority assessment unit, the third output end of the load on host computers acquiring unit 101 connect the batch and appoint
Business 203 first input end of allocation strategy generation unit, it is excellent that the output end of the load on host computers assessment unit 201 connects the host
202 second input terminal of assessment unit is first spent, 202 output end of chiller priority degree assessment unit connects the batch tasks distribution
Second input terminal of strategy generating unit 203, described batch of the output end connection of the batch tasks allocation strategy generation unit 203
Amount task distributes 3 input terminal of execution module；
The load on host computers acquiring unit 101 is sent to the master for obtaining each host current state of data center
Machine load evaluation unit 201, chiller priority degree assessment unit 202, batch tasks allocation strategy generation unit 203:
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement to task centering centre processing quantity
XQCPUj, each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud
Each available central processing unit quantity SYCPU of host current residual in data centeri, each host is current in cloud data center
Remaining free memory capacity SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, on each host
Time RWXH consumed by one step completed taski, each host last time complete task to current interval time WCJGi, it is each
A host the last time trouble duration GZCXi, each host the last time failure to current interval time GZJGi, it is each
The host last time completes the central processing unit quantity SFCPU that task is dischargedi, each host last time complete what task was discharged
Memory size SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Calculated variate-value is sent to institute for assessing each load on host computers by the load on host computers assessment unit 201
State chiller priority degree assessment unit 202:
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive execution of task of each host
Rate；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of a hosti；
A4, the equivalent expected future implementation rate for considering fault effect is set as DXZXi:
It calculatesIt is examined
Consider the equivalent expected future implementation rate of fault effect；
The chiller priority degree assessment unit (202) accepts task priority for assessing each host, and will be calculated
Variate-value is sent to the batch tasks allocation strategy generation unit (203):
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；
A is time-concerning impact factor；A is 0.5；
The batch tasks allocation strategy generation unit 203 is used to generate batch tasks allocation plan, and will be calculated
Variate-value is sent to the batch tasks distribution execution module 3:
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYWCx≥XQWCiObtain first
Collection is combined Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
The batch tasks distribution execution module 3 is for executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the use of submission task
Family；When by the corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribution
To MBjNumber host executes.
In the present embodiment, calculate
Obtain the abundant resource degree of each host ladder
CPUCYi、WCCYi、NCCYi。
In the present embodiment, the PDijThe two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
As shown in figure 3, a kind of cloud data center Batch Arrival task distribution system, the server 4 including cloud data center,
Cloud data center Batch Arrival task allocation apparatus 5, the cloud data center Batch Arrival task are provided in the server 4
Distributor 5 includes that load on host computers monitoring modular 1, " task-host " matching module 2 and batch tasks distribute execution module 3；Institute
Stating load on host computers monitoring modular 1 includes load on host computers acquiring unit 101；" task-host " matching module 2 is by load on host computers
Assessment unit 201, chiller priority degree assessment unit 202, batch tasks allocation strategy generation unit 203 form；The host is negative
The first output end for carrying acquiring unit 101 connects the input terminal of the load on host computers assessment unit 201, and the load on host computers obtains
The second output terminal of unit 101 connects 202 first input end of chiller priority degree assessment unit, and the load on host computers obtains single
The third output end of member 101 connects 203 first input end of batch tasks allocation strategy generation unit, and the load on host computers is commented
The output end for estimating unit 201 connects 202 second input terminal of chiller priority degree assessment unit, and the chiller priority degree assessment is single
First 202 output ends connect the second input terminal of the batch tasks allocation strategy generation unit 203, and the batch tasks distribute plan
Slightly the output end of generation unit 203 connects the batch tasks and distributes 3 input terminal of execution module；
The load on host computers acquiring unit 101 is sent to the master for obtaining each host current state of data center
Machine load evaluation unit 201, chiller priority degree assessment unit 202, batch tasks allocation strategy generation unit 203:
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement to task centering centre processing quantity
XQCPUj, each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud
Each available central processing unit quantity SYCPU of host current residual in data centeri, each host is current in cloud data center
Remaining free memory capacity SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, on each host
Time RWXH consumed by one step completed taski, each host last time complete task to current interval time WCJGi, it is each
A host the last time trouble duration GZCXi, each host the last time failure to current interval time GZJGi, it is each
The host last time completes the central processing unit quantity SFCPU that task is dischargedi, each host last time complete what task was discharged
Memory size SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Calculated variate-value is sent to institute for assessing each load on host computers by the load on host computers assessment unit 201
State chiller priority degree assessment unit 202:
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive execution of task of each host
Rate；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of a hosti；
A4, the equivalent expected future implementation rate for considering fault effect is set as DXZXi:
It calculatesIt is examined
Consider the equivalent expected future implementation rate of fault effect；
The chiller priority degree assessment unit 202 accepts task priority for assessing each host, and by calculated change
Magnitude is sent to the batch tasks allocation strategy generation unit 203:
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；
A is time-concerning impact factor；A is 0.5；
The batch tasks allocation strategy generation unit 203 is used to generate batch tasks allocation plan, and will be calculated
Variate-value is sent to the batch tasks distribution execution module 3:
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYWCx≥XQWCiObtain first
Collection is combined Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
The batch tasks distribution execution module 3 is for executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the use of submission task
Family；When by the corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribution
To MBjNumber host executes.
It is calculated in the present embodiment
Obtain the abundant resource degree of each host ladder
CPUCYi、WCCYi、NCCYi。
PD described in the present embodimentijThe two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that those skilled in the art without
It needs creative work according to the present invention can conceive and makes many modifications and variations.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical solution, all should be within the scope of protection determined by the claims.
Claims (9)
1. a kind of cloud data center Batch Arrival method for allocating tasks, it is characterized in that sequentially including the following steps:
Step 1: monitoring each host current state of cloud data center:
The host number in cloud data center is set as n, 0 < i≤n, n > 0；
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement XQCPU to task centering centre processing quantityj、
Each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud data center
In each available central processing unit quantity SYCPU of host current residuali, each host current residual is available in cloud data center
Memory size SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, each host last time completes
Task consumed by time RWXHi, each host last time complete task to current interval time WCJGi, each host most
Nearly primary fault duration GZCXi, each host the last time failure to current interval time GZJGi, one on each host
The central processing unit quantity SFCPU that secondary completion task is dischargedi, each host last time complete the memory size that task is discharged
SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Step 2: assessing each load on host computers:
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive implementation rate of task of each host；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each master of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of machinei；
A4, equivalent expected future implementation rate is set as DXZXi:
It calculatesObtain it is equivalent not
To be expected implementation rate；
It is assessed Step 3: accepting task priority to each host:
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；When a is
Between impact factor；
Step 4: generating batch tasks allocation plan:
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYWCx≥XQWCiObtain first collection and be combined
Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
Step 5: executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the user of submission task；As general
The corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribute to MBj
Number host executes.
2. a kind of cloud data center Batch Arrival method for allocating tasks as described in claim 1, it is characterized in that: the step 2
In, it calculates
Obtain the abundant resource degree CPUCY of each host ladderi、WCCYi、NCCYi。
3. a kind of cloud data center Batch Arrival method for allocating tasks as claimed in claim 1 or 2, it is characterized in that: the step
In rapid four, the PDijThe two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
4. a kind of cloud data center Batch Arrival task allocation apparatus, it is characterised in that: including load on host computers monitoring modular (1),
" task-host " matching module (2) and batch tasks distribution execution module (3)；The load on host computers monitoring modular (1) includes master
Machine loads acquiring unit (101)；" task-host " matching module (2) is excellent by load on host computers assessment unit (201), host
First spend assessment unit (202), batch tasks allocation strategy generation unit (203) composition；The load on host computers acquiring unit (101)
The first output end connect the input terminal of the load on host computers assessment unit (201), the load on host computers acquiring unit (101)
Second output terminal connects chiller priority degree assessment unit (202) first input end, the load on host computers acquiring unit (101)
Third output end connect batch tasks allocation strategy generation unit (203) first input end, the load on host computers assessment
The output end of unit (201) connects (202) second input terminal of chiller priority degree assessment unit, the chiller priority degree assessment
Unit (202) output end connects the second input terminal of the batch tasks allocation strategy generation unit (203), the batch tasks
The output end of allocation strategy generation unit (203) connects the batch tasks and distributes execution module (3) input terminal；
The load on host computers acquiring unit (101) is sent to the host for obtaining each host current state of data center
Load evaluation unit (201), chiller priority degree assessment unit (202), batch tasks allocation strategy generation unit (203):
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement XQCPU to task centering centre processing quantityj、
Each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud data center
In each available central processing unit quantity SYCPU of host current residuali, each host current residual is available in cloud data center
Memory size SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, each host last time completes
Task consumed by time RWXHi, each host last time complete task to current interval time WCJGi, each host most
Nearly primary fault duration GZCXi, each host the last time failure to current interval time GZJGi, one on each host
The central processing unit quantity SFCPU that secondary completion task is dischargedi, each host last time complete the memory size that task is discharged
SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Calculated variate-value is sent to described by the load on host computers assessment unit (201) for assessing each load on host computers
Chiller priority degree assessment unit (202):
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive implementation rate of task of each host；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each master of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of machinei；
A4, equivalent expected future implementation rate is set as DXZXi:
It calculatesObtain it is equivalent not
To be expected implementation rate；
The chiller priority degree assessment unit (202) accepts task priority for assessing each host, and by calculated variable
Value is sent to the batch tasks allocation strategy generation unit (203):
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；When a is
Between impact factor；
The batch tasks allocation strategy generation unit (203) is for generating batch tasks allocation plan, and by calculated change
Magnitude is sent to batch tasks distribution execution module (3):
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYNCx≥XQWCiObtain first collection and be combined
Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
Batch tasks distribution execution module (3) is for executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the user of submission task；As general
The corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribute to MBj
Number host executes.
5. a kind of cloud data center Batch Arrival task allocation apparatus as claimed in claim 4, it is characterized in that: calculating
Obtain the abundant resource degree CPUCY of each host ladderi、
WCCYi、NCCYi。
6. a kind of cloud data center Batch Arrival task allocation apparatus as described in claim 4 or 5, it is characterized in that: the PDij
The two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
7. a kind of cloud data center Batch Arrival task distribution system, the server (4) including cloud data center, it is characterized in that:
Cloud data center Batch Arrival task allocation apparatus (5), the cloud data center Batch Arrival are provided in the server (4)
Task allocation apparatus (5) includes that load on host computers monitoring modular (1), " task-host " matching module (2) and batch tasks distribution are held
Row module (3)；The load on host computers monitoring modular (1) includes load on host computers acquiring unit (101)；" task-host "
It is generated with module (2) by load on host computers assessment unit (201), chiller priority degree assessment unit (202), batch tasks allocation strategy
Unit (203) composition；First output end of the load on host computers acquiring unit (101) connects the load on host computers assessment unit
(201) second output terminal of input terminal, the load on host computers acquiring unit (101) connects the chiller priority degree assessment unit
(202) the third output end of first input end, the load on host computers acquiring unit (101) connects the batch tasks allocation strategy
Generation unit (203) first input end, the output end of the load on host computers assessment unit (201) connect the chiller priority degree and comment
Estimate (202) second input terminal of unit, chiller priority degree assessment unit (202) output end connects the batch tasks and distributes plan
Slightly generation unit (203) the second input terminal, the batch tasks allocation strategy generation unit (203) output end connection described in
Batch tasks distribute execution module (3) input terminal；
The load on host computers acquiring unit (101) is sent to the host for obtaining each host current state of data center
Load evaluation unit (201), chiller priority degree assessment unit (202), batch tasks allocation strategy generation unit (203):
It obtains cloud data center and newly reaches the quantity XDSL of task, each new requirement XQCPU to task centering centre processing quantityj、
Each requirement XQNC that capacity is newly internally deposited to taskj, each requirement XQWC that capacity is newly externally deposited to taskj, cloud data center
In each available central processing unit quantity SYCPU of host current residuali, each host current residual is available in cloud data center
Memory size SYNCi, each host current residual can use external memory capacity SYWC in cloud data centeri, each host last time completes
Task consumed by time RWXHi, each host last time complete task to current interval time WCJGi, each host most
Nearly primary fault duration GZCXi, each host the last time failure to current interval time GZJGi, one on each host
The central processing unit quantity SFCPU that secondary completion task is dischargedi, each host last time complete the memory size that task is discharged
SFNCi, each host last time complete the external memory capacity SFWC that task is dischargedi；
Calculated variate-value is sent to described by the load on host computers assessment unit (201) for assessing each load on host computers
Chiller priority degree assessment unit (202):
A1, each total abundant intensity of host residue computing resource is set as CYDi:
Calculate CYDi=CPUCYi×WCCYi×NCCYiObtain the total abundant intensity CYD of each host residue computing resourcei；
The comprehensive implementation rate of A2, the setting each host of the task is ZHXHi:
It calculatesObtain the comprehensive implementation rate of task of each host；
A3, each host for setting recent task free time influential effect the comprehensive implementation rate of task as KDXHi:
It calculatesObtain each master of recent task free time influential effect
The comprehensive implementation rate KDXH of the task of machinei；
A4, equivalent expected future implementation rate is set as DXZXi:
It calculatesObtain it is equivalent not
To be expected implementation rate；
The chiller priority degree assessment unit (202) accepts task priority for assessing each host, and by calculated variable
Value is sent to the batch tasks allocation strategy generation unit (203):
It sets each host and receives the priority of task as YXDi:
It calculatesObtain the priority that each host receives task；When a is
Between impact factor；
The batch tasks allocation strategy generation unit (203) is for generating batch tasks allocation plan, and by calculated change
Magnitude is sent to batch tasks distribution execution module (3):
B1、QjIt is combined for collection, 0 < j≤XDSL:
Calculate Qj=x | 0 < x≤n, SYCPUx≥XQCPUi, SYNCx≥XQNCi, SYNCx≥XQWCiObtain first collection and be combined
Qj；
B2、FPiFor token variable, 0 < i≤n；FPiInitial value is set as 0；
Since first in Batch Arrival task backward, the corresponding destination host labelled amount MB of the task is calculatedj:As calculated MBjWhen not being 0,It is rewritten as 1；
Batch tasks distribution execution module (3) is for executing batch tasks distribution:
When by the corresponding destination host labelled amount MB of the taskj=0, then task is refused to and is retracted the user of submission task；As general
The corresponding destination host labelled amount MB of the taskj> 0, then by the corresponding destination host labelled amount MB of the taskjDistribute to MBj
Number host executes.
8. a kind of cloud data center Batch Arrival task distribution system as claimed in claim 4, it is characterized in that: calculating
Obtain the abundant resource degree CPUCY of each host ladderi、
WCCYi、NCCYi。
9. a kind of cloud data center Batch Arrival task distribution system as claimed in claim 7 or 8, it is characterized in that: the PDij
The two-dimentional set of variables for being 0 or 1 for value；0<i≤n；0<j≤XDSL；
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811079294.6A CN109213588A (en) | 2018-09-17 | 2018-09-17 | A kind of cloud data center Batch Arrival task allocation apparatus, system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811079294.6A CN109213588A (en) | 2018-09-17 | 2018-09-17 | A kind of cloud data center Batch Arrival task allocation apparatus, system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109213588A true CN109213588A (en) | 2019-01-15 |
Family
ID=64984082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811079294.6A Pending CN109213588A (en) | 2018-09-17 | 2018-09-17 | A kind of cloud data center Batch Arrival task allocation apparatus, system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109213588A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110795255A (en) * | 2019-10-16 | 2020-02-14 | 金蝶软件（中国）有限公司 | Data batch value adjusting method and device, readable storage medium and equipment |
| CN110928689A (en) * | 2019-12-05 | 2020-03-27 | 中国人民解放军军事科学院国防科技创新研究院 | Self-adaptive resource management method and device for distributed reinforcement learning training |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104111875A (en) * | 2014-07-03 | 2014-10-22 | 重庆大学 | Device, system and method for dynamically controlling number of newly-increased tasks at cloud data center |
| CN104767806A (en) * | 2015-03-31 | 2015-07-08 | 重庆大学 | Method, device and system for backup of cloud data central task |
| CN105208119A (en) * | 2015-09-21 | 2015-12-30 | 重庆大学 | Cloud data central task allocation method, device and system |
| CN106201847A (en) * | 2016-06-30 | 2016-12-07 | 重庆大学 | Consider method for allocating tasks, device and the system of the decay of cloud platform host performance |
-
2018
- 2018-09-17 CN CN201811079294.6A patent/CN109213588A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104111875A (en) * | 2014-07-03 | 2014-10-22 | 重庆大学 | Device, system and method for dynamically controlling number of newly-increased tasks at cloud data center |
| CN104767806A (en) * | 2015-03-31 | 2015-07-08 | 重庆大学 | Method, device and system for backup of cloud data central task |
| CN105208119A (en) * | 2015-09-21 | 2015-12-30 | 重庆大学 | Cloud data central task allocation method, device and system |
| CN106201847A (en) * | 2016-06-30 | 2016-12-07 | 重庆大学 | Consider method for allocating tasks, device and the system of the decay of cloud platform host performance |
Non-Patent Citations (1)
| Title |
|---|
| 张晓磊: ""云计算独立任务及关联任务调度算法研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110795255A (en) * | 2019-10-16 | 2020-02-14 | 金蝶软件（中国）有限公司 | Data batch value adjusting method and device, readable storage medium and equipment |
| CN110928689A (en) * | 2019-12-05 | 2020-03-27 | 中国人民解放军军事科学院国防科技创新研究院 | Self-adaptive resource management method and device for distributed reinforcement learning training |
| CN110928689B (en) * | 2019-12-05 | 2020-08-25 | 中国人民解放军军事科学院国防科技创新研究院 | Self-adaptive resource management method and device for distributed reinforcement learning training |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105487930B (en) | A kind of optimizing and scheduling task method based on Hadoop | |
| CN101441580B (en) | Distributed paralleling calculation platform system and calculation task allocating method thereof | |
| CN103207814B (en) | Managing and task scheduling system and dispatching method across cluster resource of a kind of decentration | |
| Li et al. | Optimal resource provisioning for cloud computing environment | |
| Zhang et al. | Online auctions in IaaS clouds: Welfare and profit maximization with server costs | |
| Chunlin et al. | Hybrid cloud adaptive scheduling strategy for heterogeneous workloads | |
| CN103617062B (en) | The render farm Dynamic Deployment System of a kind of flexibility and method | |
| CN108845878A (en) | The big data processing method and processing device calculated based on serverless backup | |
| CN107025139A (en) | A kind of high-performance calculation Scheduling Framework based on cloud computing | |
| CN103838621B (en) | Method and system for scheduling routine work and scheduling nodes | |
| CN107003887A (en) | Overloaded cpu setting and cloud computing workload schedules mechanism | |
| CN109471705A (en) | Method, equipment and system, the computer equipment of task schedule | |
| CN103927231B (en) | The energy optimization data set distribution method that a kind of data-oriented processes | |
| CN103927229A (en) | Scheduling Mapreduce Jobs In A Cluster Of Dynamically Available Servers | |
| CN103491024B (en) | Job scheduling method and device for streaming data | |
| CN106330987A (en) | Dynamic load balancing method | |
| CN110308967A (en) | A kind of workflow cost based on mixed cloud-delay optimization method for allocating tasks | |
| CN106919449A (en) | The dispatch control method and electronic equipment of a kind of calculating task | |
| CN109040193B (en) | Mobile device cloud resource allocation method based on independent subtasks | |
| CN112219191A (en) | Self-configuration of services and servers in a data center | |
| CN109213588A (en) | A kind of cloud data center Batch Arrival task allocation apparatus, system and method | |
| CN105208119B (en) | A kind of cloud data center method for allocating tasks, device and system | |
| CN103677996B (en) | Collaboration method and system for balancing workload distribution | |
| CN107729218A (en) | A kind of system and method for monitoring processing computing resource equipment | |
| CN106201847B (en) | Consider method for allocating tasks, the device and system of the decaying of cloud platform host performance |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190115 |
|
| RJ01 | Rejection of invention patent application after publication |