CN108196939A - For the virtual machine intelligent management and device of cloud computing - Google Patents
For the virtual machine intelligent management and device of cloud computing Download PDFInfo
- Publication number
- CN108196939A CN108196939A CN201711475741.5A CN201711475741A CN108196939A CN 108196939 A CN108196939 A CN 108196939A CN 201711475741 A CN201711475741 A CN 201711475741A CN 108196939 A CN108196939 A CN 108196939A
- Authority
- CN
- China
- Prior art keywords
- queue
- request
- main
- rapid requests
- resource quantity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- 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
- G06F9/54—Interprogram communication
- G06F9/546—Message passing systems or structures, e.g. queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/54—Indexing scheme relating to G06F9/54
- G06F2209/548—Queue
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Exchange Systems With Centralized Control (AREA)
- Memory System Of A Hierarchy Structure (AREA)
Abstract
The present invention proposes a kind of virtual machine intelligent management for cloud computing, including:The main request queue of request to create and rapid requests queue, and the initialization to realize main request queue and rapid requests queue is emptied to main request queue and rapid requests queue;The request that virtual machine is sent out is received, and estimates the resource quantity called when the request performs;Obtain the resource quantity of all request calls in current main request queue, if the resource quantity of all request calls is more than preset main request queue threshold value in current main request queue, and when the resource quantity called is less than preset resource threshold during the execution of request, the tail of the queue for asking to be inserted into rapid requests queue, the otherwise request are inserted into the tail of the queue of main request queue;And it obtains and asks from team's head of the team of main request queue head and rapid requests queue respectively according to order.The present invention has the beneficial effect that:The less request of the resource quantity called during by making execution is responded as early as possible, shortens the Whole Response time.
Description
Technical field
The present invention relates to virtual machine technique field more particularly to a kind of virtual machine intelligent management for cloud computing and
Device.
Background technology
Universal with cloud computing technology, cloud computing is used by more and more mechanisms, as the alternative solution of local computing,
So as to be effectively carried out a variety of applications in a parallel fashion.Usually, associated mechanisms are achieved in that more physics independences
Server set up on the same hardware platform.It is main at present in order to facilitate the server that unified management multiple devices parameter is different
Stream scheme will be virtualized on above-mentioned hardware platform using virtual machine technique, to realize the isolation of platform, scalability and peace
Full property etc. technical indicator.After above-mentioned multiple servers are virtualized, the different type resource difference of each server is common
Various resource pools (such as hard disk pond for storing data) are formed, and passes through more virtual machines and realizes common calculating and shared money
Source, flexible allocation and efficiently utilization so as to fulfill all kinds of computing resources.
However, due to more virtual machines disorderly create a large amount of request it is shared using bottom hardware resource when, hardware money
The management and running in source must satisfy fair and efficient principle.At present, processor is normally only performed according to the request of multiple requests
Request time sequencing performs, therefore when the ratio of occupancy processor resource is excessive when some application program run, will lead
Other applications is caused to start overlong time or can not run.When serious, this will cause the operation of system to be collapsed.For example, work as
The resource quantity of some application requests is excessive and occupies for a long time, such as the processor execution time is longer, then processor needs
After first carrying out the application program, then respond the request of other applications below.This will cause occupancy resource is less below to answer
It can not be performed in time with program, reduce the overall operation efficiency of more virtual machines.
Invention content
Present invention aims to solve the deficiencies of the prior art, and provides a kind of a kind of virtual machine intelligent management sides for cloud computing
Method and device can obtain the effect for the response time for accelerating virtual machine overall applicability program.
To achieve these goals, the present invention uses following technical solution.
First, the present invention proposes a kind of virtual machine intelligent management for cloud computing, includes the following steps:Creating please
The main request queue and rapid requests queue asked, and main request queue and rapid requests queue are emptied to realize main request queue
With the initialization of rapid requests queue;The request that virtual machine is sent out is received, and estimates the number of resources called when the request performs
Amount;The resource quantity of all request calls in current main request queue is obtained, if all request calls in current main request queue
Resource quantity be more than preset main request queue threshold value, and ask execution when the resource quantity called be less than preset resource
During threshold value, the tail of the queue for asking to be inserted into rapid requests queue, the otherwise request is inserted into the tail of the queue of main request queue;With
And it obtains and asks from team's head of the team of main request queue head and rapid requests queue respectively according to order.
In the embodiment of the method for the present invention, after receiving the request that virtual machine is sent out, the data of the request are detected
It is whether normal, and abandon abnormal request.
In the embodiment of the method for the present invention, when the occupancy total resources of rapid requests queue is quick more than preset
When request queue threshold value and the occupancy total resources of main request queue are less than main request queue threshold value, positioned at rapid requests queue
Tail of the queue and request beyond preset rapid requests queue thresholds part is suggested and is inserted into the tail of the queue of main request queue.
In the embodiment of the method for the present invention, when the occupancy total resources of rapid requests queue is quick more than preset
During request queue threshold value, improve main request queue threshold value and/or reduce resource threshold.
In the embodiment of the method for the present invention, when the resource quantity of request call all in main request queue is more than in advance
If main request queue threshold value when, the resource quantity called is less than the request quilt of resource threshold when asking to perform in main request queue
It is proposed and be inserted into the tail of the queue of rapid requests queue.
Further, in the above method embodiment of the present invention, when rapid requests queue is sky in preset time period
During queue, reduce main request queue threshold value and/or improve resource threshold.
In the embodiment of the method for the present invention, after request is inserted into the tail of the queue of rapid requests queue, rapid requests
The resource quantity that queue is called when being performed according to request sorts.
In the embodiment of the method for the present invention, team's head of team's head and rapid requests queue from main request queue obtains
Request distributes the type that request is carrying out according to virtual machine.
Further, in the above method embodiment of the present invention, the resource quantity that request is called when performing is resource
Size.
Alternatively, in the above method embodiment of the present invention, the resource quantity that request is called when performing is that request accounts for
With the length of processor timeslice.
Secondly, the present invention also proposes a kind of virtual machine intelligent management apapratus for cloud computing, including with lower module:Initially
Change module, for the main request queue of request to create and rapid requests queue, and it is clear to main request queue and rapid requests queue
Sky is to realize the initialization of main request queue and rapid requests queue;Module is estimated, it please for receive that computer system sends out
It asks, and estimates the resource quantity called when the request performs;Join the team module, for obtain it is all in current main request queue please
The resource quantity of calling is sought, if the resource quantity of all request calls is more than preset main request queue in current main request queue
Threshold value, and ask execution when the resource quantity called when being less than preset resource threshold, the request is inserted into rapid requests
The tail of the queue of queue, the otherwise request are inserted into the tail of the queue of main request queue;And go out group module, for according to order respectively from
Team's head of main request queue and team's head of rapid requests queue obtain request.
In the device embodiment of the present invention, after receiving the request that virtual machine is sent out, the data of the request are detected
It is whether normal, and abandon abnormal request.
In the device embodiment of the present invention, when the resource quantity of request call all in main request queue is more than in advance
If main request queue threshold value when, the resource quantity called when asking to perform in main request queue is less than resource threshold by module of joining the team
The request of value proposes and is inserted into the tail of the queue of rapid requests queue.
In the device embodiment of the present invention, when the occupancy total resources of rapid requests queue is quick more than preset
During request queue threshold value, module of joining the team improves main request queue threshold value and/or reduces resource threshold.
Further, in the above device embodiment of the present invention, when rapid requests queue is sky in preset time period
During queue, module of joining the team reduces main request queue threshold value and/or improves resource threshold.
In the device embodiment of the present invention, when the occupancy total resources of rapid requests queue is quick more than preset
When request queue threshold value and the occupancy total resources of main request queue are less than main request queue threshold value, positioned at rapid requests queue
Tail of the queue and request beyond preset rapid requests queue thresholds part is suggested and is inserted into the tail of the queue of main request queue.
In the device embodiment of the present invention, when request is inserted by module of joining the team the tail of the queue of rapid requests queue
Afterwards, the resource quantity sort block rapid requests queue called when joining the team mould according to request execution.
In the device embodiment of the present invention, go out group module from team's head of main request queue and rapid requests queue
The request that team's head obtains distributes the type that request is carrying out according to virtual machine.
Further, in the above device embodiment of the present invention, the resource quantity that request is called when performing is resource
Size.
Alternatively, in the above device embodiment of the present invention, the resource quantity that request is called when performing is that request accounts for
With the length of processor timeslice.
Finally, the invention also discloses a kind of computer readable storage mediums, are stored thereon with computer instruction, the instruction
It is realized when being executed by processor such as the step of any one of aforementioned the method.
Beneficial effects of the present invention are:By constructing the simultaneously main request queue of maintenance request and rapid requests queue respectively,
The request for alloing the resource quantity called when performing less is responded as early as possible, so as to shorten the response time of overall request.
Description of the drawings
Fig. 1 show the method flow diagram of the virtual machine intelligent management disclosed in this invention for cloud computing;
Fig. 2 show the method flow diagram that the specific deterministic process whether request is inserted into rapid requests queue is judged in Fig. 1;
Showing for rapid requests queue is inserted into the request that Fig. 3 is shown in main request queue in one embodiment of the invention
It is intended to;
Fig. 4 show the state change schematic diagram of main request queue in another embodiment of the present invention;
Fig. 5 show the method flow diagram that the request in rapid requests queue is transferred in Fig. 4;
Fig. 6 show the state change signal that request in one embodiment of the invention is inserted into after rapid requests queue
Figure;
Fig. 7 show main request queue adjusting thresholds schematic diagram in one embodiment of the invention;
Fig. 8 show the function structure chart of the virtual machine intelligent management apapratus disclosed in this invention for cloud computing.
Specific embodiment
The technique effect of the design of the present invention, concrete structure and generation is carried out below with reference to embodiment and attached drawing clear
Chu, complete description, to be completely understood by the purpose of the present invention, scheme and effect.It should be noted that in situation about not conflicting
Under, the feature in embodiment and embodiment in the application can be combined with each other.The identical attached drawing mark used everywhere in attached drawing
Note indicates the same or similar part.
With reference to method flow diagram shown in FIG. 1, in one embodiment of the invention, for the virtual machine intelligence of cloud computing
Management method includes the following steps:The main request queue of request to create and rapid requests queue, and to main request queue and quickly
Request queue empties the initialization to realize main request queue and rapid requests queue;The request that virtual machine is sent out is received, and pre-
Estimate the resource quantity called when the request performs;The resource quantity of all request calls in current main request queue is obtained, if
The resource quantity of all request calls is more than preset main request queue threshold value in current main request queue, and during the execution of request
When the resource quantity of calling is less than preset resource threshold, the tail of the queue for asking to be inserted into rapid requests queue is otherwise described
Request is inserted into the tail of the queue of main request queue;And according to order respectively from the team of main request queue head and rapid requests queue
Team's head obtains request.Specifically, main request queue threshold value and rapid requests queue thresholds can be set in initialization.It is for example, main
Request queue threshold value can be initialized as 500K, i.e., the request in main request queue needs to occupy 500K spatial caches.At this point, it can join
Submethod flow chart shown in Fig. 2 is examined, when the spatial cache that occupancy is estimated in all requests in main request queue is less than 500K,
The tail of the queue for newly asking that main request queue will be directly inserted into that virtual machine is sent out.All requests in main request queue are according to advanced
The sequence first gone out obtains successively.When the spatial cache that occupancy is estimated in all requests in main request queue is equal to or more than 500K
When, elder generation is estimated the resource quantity called when the request performs by the new request that virtual machine is sent out.If the request performs
The resource quantity of calling is less than resource threshold (such as 20K, i.e., the spatial cache size that described request occupies when performing are less than 20K)
When, then it is inserted into the tail of the queue priority processing of rapid requests queue;Otherwise the tail of the queue of main request queue is inserted into, according to normal time
Sequence processing.Since the less request of the resource quantity of the calling request larger with the resource quantity of calling is not in same queue
It is interior, therefore shorten the response time of overall request.
With reference to the state change schematic diagram of the main request queue of embodiment illustrated in fig. 3, in the present embodiment, when main request team
When the resource quantity of all request calls is more than preset main request queue threshold value in row, it is proposed that request is held in main request queue
The resource quantity called during row is less than the request of resource threshold and is inserted into the tail of the queue of rapid requests queue.Right side side is asked in figure
Numerical value in lattice is the resource quantity called when the request performs (for example, " 10K " of A requests represents to occupy when request performs
Spatial cache size is 10K).If main request queue threshold value is 250K at this time, and the occupancy total resources of main request queue is
259K.Request of the resource quantity less than threshold value is occupied if preset resource threshold is 20K, in main request queue (" please i.e. in figure
Ask A ", " request C " and " request H " corresponding request) it will be extracted from main request queue, it is inserted into rapid requests queue
Tail of the queue so that the less request of the resource quantity that occupies can be performed in time when being performed in main request queue.
On the contrary, with reference to the state change schematic diagram of main request queue in another embodiment shown in Fig. 4, work as rapid requests
The total resources that occupies of queue is more than the occupancy total resources of preset rapid requests queue thresholds and main request queue less than master
During request queue threshold value, ask execution efficiency that will be less than main request queue in rapid requests queue.In order to improve the entirety of request
Execution efficiency with reference to submethod flow chart shown in fig. 5, positioned at the tail of the queue of rapid requests queue and can will exceed preset quick
The request (i.e. " request f " and " request g " in Fig. 4) of request queue threshold portion proposes and is inserted into the team of main request queue
Tail.
The schematic diagram of quick access request queue is inserted into reference to request shown in fig. 6, in one embodiment of the invention,
After request is inserted into the tail of the queue of quick access request queue, rapid requests queue is according to the spatial cache occupied when asking and performing
Quantity sorts.Numerical value in figure in the grid of request right side is the cache size for needing to occupy when request performs (for example, what d was asked
The spatial cache size that " 10K " represents to occupy when request performs is 10K).As shown in FIG., what is occupied when newly request f is performed is slow
It is 7K to deposit space size.If the resource quantity of all request calls is more than main request queue threshold value in main request queue at this time, f please
The tail of the queue that will first be inserted into rapid requests queue is sought, then again to rapid requests queue order.Because the money called when performing
The request of source quantity is come the front of rapid requests queue, so these requests can respond as early as possible.Further, since quickly please
Queue is asked to be maintained as orderly, so new request can be inserted into the correct position of rapid requests queue the shorter time
(in fact, the time complexity of the operation is the logarithm of current queue size).
In one embodiment of the invention, when the occupancy total resources of rapid requests queue is more than preset rapid requests
During queue thresholds, the request execution efficiency in rapid requests queue at this time will decline, even equal to or less than main request queue.For
The request for ensureing to be assigned in rapid requests queue can be performed preferentially, need to improve the door for being inserted into rapid requests queues
Sill.As shown in Figure 7, in the present embodiment, can improve main request queue threshold value (can improve to all in current main request queue
The resource quantity or higher of request call), reduce the request for being assigned to rapid requests queue.Alternatively, it quickly please be inserted into
(such as resource threshold is set as the 80% of current value) can also be improved by way of reducing resource threshold by seeking the threshold of queue,
So that access request is more difficult to be assigned to quick access request queue.
When the threshold for being inserted into rapid requests queue is improved by aforesaid way, since external cause is (such as in the unit interval
The number of requests performed is needed to reduce) or internal cause (such as be inserted into the threshold of rapid requests queue and excessively improved), make
Rapid requests queue when (such as 10 minutes) are empty queue within the preset period, be inserted into the threshold of rapid requests queue
The actual operating state to adapt to current system should appropriately be reduced.Therefore, in the above embodiment of the present invention, similarly when
In preset time period rapid requests queue be empty queue when, reduce main request queue threshold value (such as main request queue threshold value setting
80% for current value) and/or improve resource threshold (such as can improve to 2 times of current value).
In one embodiment of the invention, to further improve the operational efficiency of virtual machine, from the team of main request queue
The team of head and rapid requests queue head obtains request and distributes the type that request is carrying out according to virtual machine so that same type
Task can focus on, and the time it takes is repeatedly reassigned so as to reduce resource.
It is that request performs for weighing the resource quantity called when process performs in aforementioned several embodiments of the present invention
The spatial cache size of Shi Zhanyong.Because the numerical value can be estimated easily, and discreet value and actual value difference are less, so should
Numerical value is usually used in the cost spent when measurement process performs.Alternatively, it is another to can be used for calling when weighing request execution
Resource quantity be request occupy processor timeslice length rate.Under normal circumstances, what the holding time of processor was longer please
It asks, the resource quantity called when performing is more.Aforementioned several embodiments can be correspondingly using the length for occupying processor timeslice
Degree weighs request.
With reference to function structure chart shown in Fig. 8, in one embodiment of the invention, for the virtual machine intelligence of cloud computing
Managing device is included with lower module:Initialization module, for the main request queue of request to create and rapid requests queue, and to master
Request queue and rapid requests queue empty the initialization to realize main request queue and rapid requests queue;Module is estimated, is used
In the request that reception computer system is sent out, and estimate the resource quantity called when the request performs;It joins the team module, for obtaining
Take the resource quantity of all request calls in current main request queue, if in current main request queue all request calls resource
Quantity is more than preset main request queue threshold value, and the resource quantity called during the execution of request is less than preset resource threshold
When, the tail of the queue for asking to be inserted into rapid requests queue, the otherwise request is inserted into the tail of the queue of main request queue;And go out
Team's module is asked for being obtained respectively from team's head of the team of main request queue head and rapid requests queue according to order.Specifically,
Main request queue threshold value and rapid requests queue thresholds can be set in initialization.For example, main request queue threshold value can initialize
For 500K, i.e., the request in main request queue needs to occupy 500K spatial caches.At this point, when the entirety request in main request queue
When estimating the spatial cache of occupancy and being less than 500K, the tail of the queue for newly asking that main request queue will be directly inserted into that virtual machine is sent out.
All requests in main request queue obtain successively according to the sequence of first in first out.When all requests in main request queue are estimated
When the spatial cache of occupancy is equal to or more than 500K, the new request that virtual machine is sent out is called when elder generation is estimated the request execution
Resource quantity.If the resource quantity called is less than resource threshold (such as 20K, i.e., described request performs when the request performs
The spatial cache size of Shi Zhanyong is less than 20K) when, then it is inserted into the tail of the queue priority processing of rapid requests queue;Otherwise it is inserted into
To the tail of the queue of main request queue, handled according to regular turn.Due to the less request of the resource quantity of calling and the resource called
The larger request of quantity be not in same queue, therefore shorten overall request response time.
With reference to the state change schematic diagram of the main request queue of embodiment illustrated in fig. 3, in the present embodiment, when main request team
When the resource quantity of all request calls is more than preset main request queue threshold value in row, module of joining the team is proposed that main request queue
The resource quantity that middle request is called when performing is less than the request of resource threshold and is inserted into the tail of the queue of rapid requests queue.It please in figure
Numerical value on the right side of asking in grid is the resource quantity called when the request performs (for example, " 10K " of A requests represents request execution
The spatial cache size of Shi Zhanyong is 10K).If main request queue threshold value is 250K at this time, and the occupancy resource of main request queue
Total amount is 259K.If preset resource threshold is 20K, module of joining the team will occupy resource quantity and be less than threshold value in main request queue
Request (" request A ", " request C " and " request H " corresponding request i.e. in figure) will be extracted from main request queue, insertion
To the tail of the queue of rapid requests queue so that the less request of the resource quantity that occupies can be timely when being performed in main request queue
It performs.
On the contrary, with reference to the state change schematic diagram of main request queue in another embodiment shown in Fig. 4, work as rapid requests
The total resources that occupies of queue is more than the occupancy total resources of preset rapid requests queue thresholds and main request queue less than master
During request queue threshold value, ask execution efficiency that will be less than main request queue in rapid requests queue.In order to improve the entirety of request
Execution efficiency, with reference to submethod flow chart shown in fig. 5, module of joining the team positioned at the tail of the queue of rapid requests queue and can will exceed pre-
If rapid requests queue thresholds part request (i.e. " request f " and " request g " in Fig. 4) proposition and be inserted into main request team
The tail of the queue of row.
The schematic diagram of quick access request queue is inserted into reference to request shown in fig. 6, in one embodiment of the invention,
After request is inserted into the tail of the queue of quick access request queue, rapid requests queue is according to the spatial cache occupied when asking and performing
Quantity sorts.Numerical value in figure in the grid of request right side is the cache size for needing to occupy when request performs (for example, what d was asked
The spatial cache size that " 10K " represents to occupy when request performs is 10K).As shown in FIG., what is occupied when newly request f is performed is slow
It is 7K to deposit space size.If the resource quantity of all request calls is more than main request queue threshold value in main request queue at this time, enter
F requests are first inserted into the tail of the queue of rapid requests queue by team's module, then again to rapid requests queue order.Because tune when performing
The request of resource quantity is come the front of rapid requests queue, so these requests can respond as early as possible.Further, since
Rapid requests queue be maintained as it is orderly, so new request can be inserted into rapid requests queue just the shorter time
True position (in fact, the time complexity of the operation is the logarithm of current queue size).
In one embodiment of the invention, when the occupancy total resources of rapid requests queue is more than preset rapid requests
During queue thresholds, the request execution efficiency in rapid requests queue at this time will decline, even equal to or less than main request queue.For
The request for ensureing to be assigned in rapid requests queue can be performed preferentially, need to improve the door for being inserted into rapid requests queues
Sill.As shown in FIG., in the present embodiment, module of joining the team improves main request queue threshold value, and (dotted portion in figure can be improved to working as
The resource quantity or higher of all request calls in preceding main request queue), reduce the request for being assigned to rapid requests queue.It can replace
Dai Di, being inserted into the threshold of rapid requests queue can also improve that (such as resource threshold is set by way of reducing resource threshold
It is set to the 80% of current value) so that access request is more difficult to be assigned to quick access request queue.
When the threshold for being inserted into rapid requests queue is improved by aforesaid way, since external cause is (such as in the unit interval
The number of requests performed is needed to reduce) or internal cause (such as be inserted into the threshold of rapid requests queue and excessively improved), make
Rapid requests queue when (such as 10 minutes) are empty queue within the preset period, be inserted into the threshold of rapid requests queue
The actual operating state to adapt to current system should appropriately be reduced.Therefore, in the above embodiment of the present invention, similarly when
When rapid requests queue is empty queue in preset time period, module of joining the team can reduce main request queue threshold value (such as main request team
Row threshold value is set as the 80% of current value) and/or raising resource threshold (such as can improve to 2 times of current value).
In one embodiment of the invention, to further improve the operational efficiency of virtual machine, from the team of main request queue
The team of head and rapid requests queue head obtains request and distributes the type that request is carrying out according to virtual machine so that same type
Task can focus on, and the time it takes is repeatedly reassigned so as to reduce resource.
It is that request performs for weighing the resource quantity called when process performs in aforementioned several embodiments of the present invention
The spatial cache size of Shi Zhanyong.Because the numerical value can be estimated easily, and discreet value and actual value difference are less, so should
Numerical value is usually used in the cost spent when measurement process performs.Alternatively, it is another to can be used for calling when weighing request execution
Resource quantity be request occupy processor timeslice length rate.Under normal circumstances, what the holding time of processor was longer please
It asks, the resource quantity called when performing is more.Aforementioned several embodiments can be correspondingly using the length for occupying processor timeslice
Degree weighs request.
Although description of the invention is quite detailed and especially several embodiments are described, it is not
Any of these details or embodiment or any specific embodiments are intended to be limited to, but it is by reference to appended that should be considered as
Claim considers that the prior art provides the possibility explanation of broad sense for these claims, so as to effectively cover the present invention
Preset range.In addition, with the foreseeable embodiment of inventor, present invention is described above, its purpose is to be provided with
Description, and those unsubstantiality changes to the present invention still unforeseen at present can still represent the equivalent modifications of the present invention.
Claims (10)
1. a kind of virtual machine intelligent management for cloud computing, which is characterized in that include the following steps:
The main request queue of request to create and rapid requests queue, and main request queue and rapid requests queue are emptied to realize
Main request queue and the initialization of rapid requests queue;
The request that virtual machine is sent out is received, and estimates the resource quantity called when the request performs;
The resource quantity of all request calls in current main request queue is obtained, if all request calls in current main request queue
Resource quantity be more than preset main request queue threshold value, and ask execution when the resource quantity called be less than preset resource
During threshold value, the tail of the queue for asking to be inserted into rapid requests queue, the otherwise request is inserted into the tail of the queue of main request queue;With
And
According to order request is obtained from the team of main request queue head and team's head of rapid requests queue respectively.
2. method according to claim 1, which is characterized in that after receiving the request that virtual machine is sent out, estimate the request
Whether data are normal, and abandon abnormal request.
3. method according to claim 1, which is characterized in that when the resource that main request queue occupies is more than preset main request
During queue thresholds, the request for the resource quantity called when performing being asked to be less than resource threshold in main request queue is suggested and is inserted into
To the tail of the queue of rapid requests queue.
4. method according to claim 1, which is characterized in that when the resource that rapid requests queue occupies is quick more than preset
During request queue threshold value, improve main request queue threshold value and/or reduce resource threshold.
5. method according to claim 1, which is characterized in that when the occupancy total resources of rapid requests queue is more than preset
When rapid requests queue thresholds and the occupancy total resources of main request queue are less than main request queue threshold value, positioned at rapid requests team
The tail of the queue of row and request beyond preset rapid requests queue thresholds part is suggested and is inserted into the tail of the queue of main request queue.
6. method according to claim 1, which is characterized in that after request is inserted into the tail of the queue of rapid requests queue, quickly
The resource quantity that request queue is called when being performed according to request sorts.
7. according to the method any in claim 1 to 6, which is characterized in that the resource quantity that request is called when performing is money
The size in source.
8. according to the method any in claim 1 to 6, which is characterized in that the resource quantity that request is called when performing is please
Seek the length for occupying processor timeslice.
9. a kind of virtual machine intelligent management apapratus for cloud computing, which is characterized in that including with lower module:
Initialization module for the main request queue of request to create and rapid requests queue, and to main request queue and quickly please
Queue is asked to empty to realize the initialization of main request queue and rapid requests queue;
Module is estimated, for receiving the request that computer system is sent out, and estimates the resource quantity called when the request performs;
It joins the team module, for obtaining the resource quantity of all request calls in current main request queue, if current main request queue
The resource quantity of interior entirety request call is more than preset main request queue threshold value, and the resource quantity called during the execution of request
During less than preset resource threshold, the tail of the queue for asking to be inserted into rapid requests queue, the otherwise request is inserted into main ask
Seek the tail of the queue of queue;And
Go out group module, asked for being obtained respectively from team's head of the team of main request queue head and rapid requests queue according to order.
10. a kind of computer readable storage medium, is stored thereon with computer instruction, it is characterised in that the instruction is held by processor
It is realized during row such as the step of method described in any item of the claim 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711475741.5A CN108196939B (en) | 2017-12-29 | 2017-12-29 | Intelligent virtual machine management method and device for cloud computing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711475741.5A CN108196939B (en) | 2017-12-29 | 2017-12-29 | Intelligent virtual machine management method and device for cloud computing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108196939A true CN108196939A (en) | 2018-06-22 |
CN108196939B CN108196939B (en) | 2022-02-18 |
Family
ID=62586236
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711475741.5A Active CN108196939B (en) | 2017-12-29 | 2017-12-29 | Intelligent virtual machine management method and device for cloud computing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108196939B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109213561A (en) * | 2018-09-14 | 2019-01-15 | 珠海国芯云科技有限公司 | The equipment scheduling method and device of virtual desktop based on container |
CN109871262A (en) * | 2019-02-28 | 2019-06-11 | 北京隆普智能科技有限公司 | A kind of method and its creating device of virtual machine creating |
CN109992357A (en) * | 2019-04-10 | 2019-07-09 | 北京隆普智能科技有限公司 | It is a kind of to execute the method and its system that conflict is avoided when change for virtual machine |
CN110442431A (en) * | 2019-08-12 | 2019-11-12 | 安徽赛福贝特信息技术有限公司 | The creation method of virtual machine in a kind of cloud computing system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100094852A1 (en) * | 2008-10-14 | 2010-04-15 | Chetan Kumar Gupta | Scheduling queries using a stretch metric |
CN104202261A (en) * | 2014-08-27 | 2014-12-10 | 华为技术有限公司 | Service request processing method and device |
CN104199739A (en) * | 2014-08-26 | 2014-12-10 | 浪潮(北京)电子信息产业有限公司 | Speculation type Hadoop scheduling method based on load balancing |
CN105448006A (en) * | 2015-12-30 | 2016-03-30 | 武汉邮电科学研究院 | Intelligent supermarket cashier system and method based on mobile payment and IOT (Internet of Things) |
CN105991588A (en) * | 2015-02-13 | 2016-10-05 | 华为技术有限公司 | ethod and apparatus for resisting message attack |
CN106470169A (en) * | 2015-08-19 | 2017-03-01 | 阿里巴巴集团控股有限公司 | A kind of service request method of adjustment and equipment |
CN107018175A (en) * | 2017-01-11 | 2017-08-04 | 杨立群 | The dispatching method and device of mobile cloud computing platform |
-
2017
- 2017-12-29 CN CN201711475741.5A patent/CN108196939B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100094852A1 (en) * | 2008-10-14 | 2010-04-15 | Chetan Kumar Gupta | Scheduling queries using a stretch metric |
CN104199739A (en) * | 2014-08-26 | 2014-12-10 | 浪潮(北京)电子信息产业有限公司 | Speculation type Hadoop scheduling method based on load balancing |
CN104202261A (en) * | 2014-08-27 | 2014-12-10 | 华为技术有限公司 | Service request processing method and device |
CN105991588A (en) * | 2015-02-13 | 2016-10-05 | 华为技术有限公司 | ethod and apparatus for resisting message attack |
CN106470169A (en) * | 2015-08-19 | 2017-03-01 | 阿里巴巴集团控股有限公司 | A kind of service request method of adjustment and equipment |
CN105448006A (en) * | 2015-12-30 | 2016-03-30 | 武汉邮电科学研究院 | Intelligent supermarket cashier system and method based on mobile payment and IOT (Internet of Things) |
CN107018175A (en) * | 2017-01-11 | 2017-08-04 | 杨立群 | The dispatching method and device of mobile cloud computing platform |
Non-Patent Citations (2)
Title |
---|
XIANLIANG JIANG等: ""LRURC: A Low Complexity and Approximate Fair Active Queue Management Algorithm for Choking Non-Adaptive Flows"", 《IEEE COMMUNICATIONS LETTERS》 * |
陆婷等: ""基于HBase的交通流数据实时存储系统"", 《计算机应用》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109213561A (en) * | 2018-09-14 | 2019-01-15 | 珠海国芯云科技有限公司 | The equipment scheduling method and device of virtual desktop based on container |
CN109871262A (en) * | 2019-02-28 | 2019-06-11 | 北京隆普智能科技有限公司 | A kind of method and its creating device of virtual machine creating |
CN109992357A (en) * | 2019-04-10 | 2019-07-09 | 北京隆普智能科技有限公司 | It is a kind of to execute the method and its system that conflict is avoided when change for virtual machine |
CN109992357B (en) * | 2019-04-10 | 2022-01-14 | 中航金网(北京)电子商务有限公司 | Method and system for avoiding conflict when executing change to virtual machine |
CN110442431A (en) * | 2019-08-12 | 2019-11-12 | 安徽赛福贝特信息技术有限公司 | The creation method of virtual machine in a kind of cloud computing system |
Also Published As
Publication number | Publication date |
---|---|
CN108196939B (en) | 2022-02-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108196939A (en) | For the virtual machine intelligent management and device of cloud computing | |
CN110287003B (en) | Resource management method and management system | |
Grandl et al. | Multi-resource packing for cluster schedulers | |
CN110413412B (en) | GPU (graphics processing Unit) cluster resource allocation method and device | |
CN109992366B (en) | Task scheduling method and task scheduling device | |
CN103927225A (en) | Multi-core framework Internet information processing and optimizing method | |
CN109697122A (en) | Task processing method, equipment and computer storage medium | |
US10089155B2 (en) | Power aware work stealing | |
CN106095940A (en) | A kind of data migration method of task based access control load | |
CN111506434B (en) | Task processing method and device and computer readable storage medium | |
CN104243617A (en) | Task scheduling method and system facing mixed load in heterogeneous cluster | |
US20180039520A1 (en) | Methods and Nodes for Scheduling Data Processing | |
CN108897622A (en) | A kind of dispatching method and relevant apparatus of task run | |
CN108694083B (en) | Data processing method and device for server | |
CN112015765A (en) | Spark cache elimination method and system based on cache value | |
CN108304254A (en) | Quick virtual machine process dispatch control method and device | |
CN110162398B (en) | Scheduling method and device of disease analysis model and terminal equipment | |
CN114564302A (en) | GPU resource allocation method, system, device and medium | |
CN114721818A (en) | Kubernetes cluster-based GPU time-sharing method and system | |
US20210390405A1 (en) | Microservice-based training systems in heterogeneous graphic processor unit (gpu) cluster and operating method thereof | |
CN111367655B (en) | Method, system and storage medium for GPU resource scheduling in cloud computing environment | |
CN107423114B (en) | Virtual machine dynamic migration method based on service classification | |
CN107528871A (en) | Data analysis in storage system | |
CN104866370A (en) | Dynamic time slice dispatching method and system for parallel application under cloud computing environment | |
CN110175078B (en) | Service processing 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |