CN111078402A - Resource pool system capable of rapidly providing experimental environment - Google Patents
Resource pool system capable of rapidly providing experimental environment Download PDFInfo
- Publication number
- CN111078402A CN111078402A CN201911221886.1A CN201911221886A CN111078402A CN 111078402 A CN111078402 A CN 111078402A CN 201911221886 A CN201911221886 A CN 201911221886A CN 111078402 A CN111078402 A CN 111078402A
- Authority
- CN
- China
- Prior art keywords
- resource
- resource pool
- mirror image
- scheduling
- mirror
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- 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
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5011—Pool
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5022—Workload threshold
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/508—Monitor
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a resource pool system for rapidly providing an experimental environment, which comprises: the system comprises a resource pool configuration module, a heat mirror management module, a resource scheduling module, a resource counting module and a cloud driving module; the resource pool configuration module is used for configuring a resource pool system and controlling the resource pool to be started and closed; the configuration of the resource pool comprises the total size setting of the resource pool, the minimum idle resource number setting of the resource pool, the hot spot mirror image statistical method setting and the resource scheduling method setting; the heat mirror image management module is used for counting the heat mirror images according to a configured heat mirror image counting method; the resource scheduling module is used for performing dynamic telescopic resource allocation on the hotspot mirror images counted according to the configured resource scheduling method and the minimum idle resource number of the resource pool; the resource counting module is used for counting resource data of each hotspot mirror image of the resource pool; the cloud driving module is used for butting a cloud computing platform supported by the bottom layer and calling an interface of the cloud computing platform.
Description
Technical Field
The invention relates to the field of cloud computing platforms, in particular to a resource pool system for rapidly providing an experimental environment.
Background
A service-based cloud computing platform is an important form of information infrastructure in the Internet age, and is the latest form of development of high-performance and distributed computing. The method provides high-performance and low-cost calculation and data service in a new business mode, and supports various informationized applications. The method is a novel informatization mode which is based on virtualization, takes service as a characteristic and takes on-demand use as a business mode. After birth, the cloud computing concept quickly draws strong attention of governments, industrial circles and academic circles, and popularization and application of the cloud computing concept are out of the way. And various large enterprises develop cloud computing products and services in a dispute, and strive to seize the highest point of the cloud computing application market.
At present, a practical training cloud system exists in the market, a back-end structure of a product can be supported by a cloud computing platform, computing service is provided for a practical training cloud through the cloud computing platform, the practical training cloud system can be single-node cloud host service, can also be multi-node resource arrangement service, and meanwhile, practical training environment isolation of different users of the practical training cloud can also be achieved. The cloud computing platform is a good choice for training cloud rear-end support.
In the using process, the practical training cloud system starts a practical training environment, and has to face the problem of time consumption for creating a cloud host even if a cloud computing platform is adopted; and the problems of the overall architecture and performance of the system are examined when the cloud host is created and started to be completely prepared.
Disclosure of Invention
The invention provides a resource pool system for rapidly providing an experimental environment, which is characterized in that when a target system needs the experimental environment, resources of corresponding types are directly obtained from the resource pool system to serve as the experimental environment, the rapid response capability of the target system to the experimental environment is improved, and the time for starting the experimental environment is shortened.
In order to solve the above technical problem, an embodiment of the present invention provides a resource pool system for quickly providing an experimental environment, including: the system comprises a resource pool configuration module, a heat mirror management module, a resource scheduling module, a resource counting module and a cloud driving module;
the resource pool configuration module is used for configuring a resource pool system and controlling the resource pool to be started and closed; the configuration of the resource pool comprises the total size setting of the resource pool, the minimum idle resource number setting of the resource pool, the hot spot mirror image statistical method setting and the resource scheduling method setting;
the heat mirror image management module is used for counting the heat mirror images according to a configured heat mirror image counting method;
the resource scheduling module is used for performing dynamic telescopic allocation on the resources for the counted hotspot mirror images according to the configured resource scheduling method and the minimum idle resource number of the resource pool;
the resource counting module is used for counting resource data of each hotspot mirror image of the resource pool;
the cloud driving module is used for butting a cloud computing platform supported by the bottom layer and calling an interface of the cloud computing platform.
As a preferred scheme, the heat image management module performs heat image statistics by a full image statistical method, where the full image statistical method specifically includes:
counting the previous n mirror image historical use records;
sorting the n records according to mirror names, installing the records, and sequencing the records into X according to the use times;
classifying the X according to the same use times of the mirror images and sequencing the X into Y according to the times;
sorting the n records according to the mirror image service time and removing the duplicate into Z;
allocating the name times to the corresponding mirror image of the Y according to the times from large to small;
and recording the weight of the corresponding mirror image, wherein the using times of the mirror image is the weight of the mirror image.
Preferably, before recording the weight of the corresponding image, the method further includes: when the use times of a plurality of images are the same, ranking is carried out according to the order of the images in Z.
As a preferred scheme, the heat image management module performs heat image statistics by a quantitative image statistics method, and the quantitative image statistics method specifically includes:
counting the previous n mirror image historical use records;
sorting the n records according to mirror names, installing the records, and sequencing the records into X according to the use times;
classifying the X according to the same use times of the mirror images and sequencing the X into Y according to the times;
allocating the name times to the corresponding mirror image of the Y according to the times from large to small;
and recording the weight of the corresponding mirror image, wherein the using times of the mirror image is the weight of the mirror image.
Preferably, before recording the weight of the corresponding image, the method further includes: when the use times of a plurality of mirror images are the same, counting the top 2n mirror images in the historical use record, sequencing the plurality of mirror images according to the times and distributing the names.
As a preferred scheme, the resource scheduling module is configured to perform a step of dynamically allocating resources in a flexible manner, and specifically includes:
acquiring a configured resource scheduling mode, and executing a scheduling task according to the resource scheduling mode;
acquiring a hot mirror noun and weight, and acquiring the minimum idle number of a resource pool;
performing heat mirror image idle resource number distribution according to the ranking, and taking 1 when the distributed number is less than 1; when the distributed number is an integer, taking the integer; when the distributed number is a non-integer greater than 1, taking an integer part; and the cloud driving module creates idle resources for each mirror image according to the number of the idle resources allocated to each mirror image.
As a preferred scheme, the resource scheduling manner is a timing scheduling manner, and the timing scheduling manner includes: and setting a scheduling time point, and triggering a resource scheduling task when the system time reaches the scheduling time point.
As a preferred scheme, the resource scheduling manner is a real-time scheduling manner, and the real-time scheduling manner includes: and monitoring the change of the hot mirror image in real time, and triggering a resource scheduling task when the judgment is changed.
As a preferred scheme, the resource scheduling manner is a manual scheduling manner, and the manual scheduling manner includes: and directly triggering the resource scheduling task.
As a preferred scheme, the resource scheduling manner is a low load scheduling manner, and the low load scheduling manner includes: and monitoring the load condition of the cloud computing platform in real time, and directly triggering the resource scheduling task when the load of the cloud computing platform is detected and judged to be lower than a set threshold value.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
the resource pool system is configured, the heat image is counted according to a configured heat image counting method, the resources are dynamically and telescopically distributed according to a configured resource scheduling method, and driving calling is carried out; when the target system needs the experiment environment, resources of corresponding types are directly obtained from the resource pool system to serve as the experiment environment, the quick response capability of the target system to the experiment environment is improved, and the time for starting the experiment environment is shortened.
Drawings
FIG. 1: the structure diagram of the resource pool system in the embodiment of the invention is shown;
FIG. 2: an execution flow chart of a resource pool system for rapidly providing an experimental environment is provided;
FIG. 3: a schematic flow chart of hot spot mirror image statistics is shown in the embodiment of the invention;
FIG. 4: a flow chart of performing full-scale statistics on hotspot images is provided in the embodiment of the invention;
FIG. 5: a flow chart for quantitatively counting hot spot mirror images is carried out in the embodiment of the invention;
FIG. 6: a schematic flow chart of resource allocation in the embodiment of the present invention is shown.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 to 6, a resource pool system for rapidly providing an experimental environment according to a preferred embodiment of the present invention includes: the system comprises a resource pool configuration submodule, a hotspot mirror image management module, a resource scheduling module, a resource counting module and a cloud driving module.
The resource pool configuration module is used for configuring the resource pool system and enabling and closing the resource pool function. The configuration of the resource pool comprises the total size setting of the resource pool, the minimum idle resource number setting of the resource pool, the hot spot mirror image statistical method setting and the resource scheduling method setting. The hotspot mirror image statistical method and the resource scheduling method are set to be high-level optional configurations.
The total size of the resource pool specifies the total resource quantity of the resource pool, and can be determined according to physical resources of the cloud computing platform and actual requirements of the practical training cloud. The number of free resources in the resource pool plus the number of used resources in the resource pool cannot exceed the total number of resources in the resource pool.
The hot image management module is used for carrying out intelligent statistics on the hot images according to the configured hot image statistical method and realizing manual management on the hot images.
And the resource scheduling module is used for performing dynamic telescopic allocation on the resources for the counted hotspot images according to the configured resource scheduling method and the minimum idle resource number of the resource pool.
The resource counting module is used for counting resources of hot spot mirror images of the resource pool, wherein the resources comprise resources which are used and idle resources.
The cloud driving module is used for butting a cloud computing platform supported by a bottom layer, and realizing interface calling of the cloud computing platform, including operations of obtaining, creating, deleting and the like of various resources of the cloud platform.
The hotspot mirror image statistical method comprises the following steps: 1. carrying out quantitative mirror image statistics; 2. and (5) carrying out full-scale mirror image statistics.
The quantitative mirror image statistics means that according to the latest n records of mirror image use, the first m mirror images with the largest use times or heat degrees are counted as the hot spot mirror images, and the use times are used as the weight of each hot spot mirror image. And carrying out heat sorting according to the using times of the mirror images. If the use times of the plurality of images are equal, the latest 2 x n pieces of data are continuously searched, and the plurality of images are reordered to determine the image heat. Looking up in this way until the first m hot spot images are ejected.
The total mirror image statistics means that all mirror images are counted according to the latest n records used by the mirror images, all the mirror images are used as hotspot mirror images, and the use times m of each hotspot mirror image are used as the weight of each hotspot mirror image. And carrying out heat sorting according to the using times of the mirror images. If the plurality of images are used an equal number of times, the last used image of the plurality of images is ranked higher in popularity than the non-last used image.
The weight of each hotspot image and the ranking rank determine the number of idle resources of the hotspot image.
When allocating resources for each hot mirror image from the minimum idle resource number, allocating the resource number preferentially for the mirror image with the top rank; when the number of the resources distributed according to the weight is less than 1, taking 1 as the number of the distributed resources; and when the number of the resources allocated according to the weight is a non-integer greater than 1, taking the integer part as the number of the allocated resources.
The resource scheduling method comprises the following steps: 1. low load scheduling; 2. scheduling in real time; 3. scheduling at fixed time; 4. and (4) manual scheduling. The low-load scheduling refers to resource scheduling performed when the load of the cloud computing platform is low or is lower than a certain threshold. The real-time scheduling refers to monitoring a resource pool in real time, and scheduling the resource pool immediately when a change needs to be performed. The timing scheduling refers to manually configuring scheduling time, and performing resource scheduling each time when the time point is reached. The manual scheduling refers to an operation that can perform resource scheduling on a management interface according to a change of a resource demand.
When the practical training cloud user needs an experimental environment, whether the experimental environment type is in the resource pool or not can be judged preferentially, and if the experimental environment type is in the resource pool, one resource is selected from a plurality of resources of the experimental type through the resource scheduling submodule to be scheduled to the practical training cloud user.
And if the resource is not in the resource pool, scheduling the cloud driving sub-module of the resource pool to create the resource, and then scheduling the resource to the user. After the scheduling is completed, the requirements of the user are recorded, so that the hot spot mirror is counted by the hot spot mirror counting submodule by using the requirement records. And after the resources are allocated to the users, adding 1 to the resource usage number of the corresponding hotspot image, and subtracting 1 from the number of the idle resources. And when the practical training cloud is released to the experimental environment, the resource scheduling submodule deletes the corresponding used resource.
The present invention will be described in detail with reference to specific examples.
Fig. 1 shows a structure diagram of a resource pool system according to an embodiment of the present invention. In the technical scheme, a practical training cloud system is used as a target system for research. On the whole framework, the training cloud is obtained through the resource pool system, the existing cloud platform example is obtained, and the cloud platform example is used as the experimental environment of the training cloud for users to use.
Specifically, the practical training cloud can realize necessary configuration of the resource pool through a resource pool configuration submodule of the resource pool system, and after the configuration is completed, the resource pool configuration submodule can trigger a hot mirror image statistical task according to a set mirror image statistical method. After the hot image statistics is completed, the hot image statistics submodule triggers a resource scheduling task according to the set resource scheduling method to complete the establishment of the resource pool.
Specifically, the training cloud can manage the heat mirror images through statistics of a heat mirror image statistics submodule of the resource pool system, wherein the management includes acquiring all the heat mirror images, checking heat mirror image weights and manually adding the heat mirror images. When the counted hot mirror image is changed, the hot mirror image counting submodule triggers a resource scheduling task according to the set resource scheduling method to complete the establishment of the resource pool.
Specifically, the training cloud may obtain all resource information of the resource pool through the resource statistics submodule of the resource pool system, including resource information being used by each hotspot mirror image and idle resource information.
Fig. 2 shows a detailed flow chart of the method according to the embodiment of the invention. Firstly, the training cloud opens a resource pool, and resource pool configuration is carried out through a resource pool configuration submodule, wherein the resource pool configuration comprises the total number of resource pool resources and the minimum number of idle resources of the resource pool, and advanced optional configuration comprises a hot spot mirror image statistical method and a resource scheduling method; secondly, the hotspot mirror image counting submodule counts the hotspot mirror images and the weights thereof through the use records of the historical mirror images; then, the resource scheduling submodule performs idle resource allocation of each hotspot mirror image through resource scheduling; and finally, idle resources of each hotspot image created by the resource pool system are provided for the practical training cloud as an experimental environment.
Fig. 3 is a schematic view illustrating a statistical process of the hot spot mirroring in step S1 in fig. 2. Firstly, acquiring the historical use records of the first n (n can be any integer) mirror images by a hotspot mirror image statistics submodule; secondly, acquiring a mirror image statistical method, wherein the two methods are adopted, the first method is a full-quantity mirror image statistical method, and the second method is a quantitative statistical method; then, a specified mirror image statistical method is adopted to count hot spot mirror images and corresponding ranking and weight, the ranking determines the sequence of distributing idle resources, the ranking is distributed first, and the weight determines the number of the idle resources distributed by the hot spot mirror images.
FIG. 4 shows the full-scale statistical hotspot mirroring flow diagram indicated in FIG. 3. The core idea of the total statistical hotspot mirror image method is to classify all the n mirror images m according to the use records of near n (self-defined integers larger than 0) mirror images, and take all the m mirror images as hotspot mirror images. The number of times the mirror is used in these n records (from large to small) ranks the names of the mirrors from 1-m. If the number of times of using a plurality of images is the same, the images are sorted according to the using time with near and far, and the closer to the current time, the more advanced the ranking.
Specifically, when a full-quantity mirror image statistical method is adopted and the hotspot mirror image statistics needs to be executed, firstly, a hotspot mirror image statistical submodule can perform statistics on the previous n mirror image historical use records; secondly, classifying and installing the n records according to mirror names, and sequencing the use times of the n records into X; then, classifying the X according to the same use times of the mirror images and sequencing the X into Y according to the times; then, sorting the n records according to the mirror image service time and removing the duplicate into Z; then, assigning the Y to the corresponding mirror images from 1 to m (the first n mirror images are m in total) according to the number of times from large to small; when the use times of a plurality of mirror images are the same, ranking the mirror images according to the sequence of the mirror images in Z; and finally, recording the weight of the corresponding mirror image, wherein the using times of the mirror image is the weight of the mirror image, and completing the total statistics of the hotspot mirror images.
FIG. 5 shows the quantitative statistical hotspot mirroring flow diagram indicated in FIG. 3. The core idea of the quantitative statistical hotspot mirror image method is to classify all n mirror images m according to the use records of near n (the self-defined integer larger than 0) mirror images, sort the m mirror images according to the use times, and then take the first a (the self-defined integer larger than 0) mirror images as hotspot mirror images. The number of times the mirror is used in these n records (from large to small) ranks the name of the mirror from 1-a. If the times of using a plurality of images are the same, continuing to take 2 x n image records, sorting the images from large to small according to the times of using the images in the 2 x n records, wherein the earlier the sorting is, the earlier the ranking is, if the times of using the old images are the same, continuing to recursively determine the image name in the mode.
Specifically, when a quantitative mirror image statistical method is adopted and hot spot mirror image statistics needs to be executed, firstly, a hot spot mirror image statistical submodule counts n previous mirror image historical use records; secondly, classifying and installing the n records according to mirror names, and sequencing the use times of the n records into X; then, classifying the X according to the same use times of the mirror images and sequencing the X into Y according to the times; then, assigning the Y to the corresponding mirror images from 1 to a (m mirror images are shared in the first n pieces) according to the times from large to small; when the use times of a plurality of mirror images are the same, counting (i +1) × n before the mirror image historical use record, sequencing the mirror images according to the times and distributing the names; and finally, recording the weight of the corresponding mirror image, wherein the using times of the mirror image is the weight of the mirror image, and completing the total statistics of the hotspot mirror images.
Fig. 6 is a schematic flow chart illustrating resource allocation in step S2 in fig. 2. The resource scheduling can be set with four scheduling modes, wherein the first mode is timing scheduling, the second mode is real-time scheduling, the third mode is manual scheduling, and the fourth mode is low-load scheduling. When the timing scheduling is adopted, a certain time point needs to be configured to the resource scheduling module in advance, and when the system time reaches the time point, a resource scheduling task is triggered; when real-time scheduling is adopted, the change of the hot mirror image is monitored in real time, wherein the change comprises whether the hot mirror image is changed or not, whether the number of idle resources of the hot mirror image is changed or not and the like, when the change is detected, a resource scheduling task is triggered, and the system performance is possibly influenced relatively in the mode; when manual scheduling is adopted, a resource scheduling task is directly triggered, and the mode can be compatible with other three modes; when low-load scheduling is adopted, the load condition of the cloud computing platform needs to be monitored in real time, and when the load of the cloud computing platform is detected to be lower than a set threshold value, a resource scheduling task is directly triggered.
When the resource scheduling module executes the scheduling task, all the popularity mirror names, the weights and the minimum idle number of the resource pool are obtained firstly. Then, carrying out idle resource number distribution according to the heat mirror image ranking and the weight, and taking 1 when the distributed number is less than 1; when the distributed number is an integer, taking the integer; and when the distributed number is a non-integer larger than the non-integer, taking an integer. And finally, the cloud driving module creates a corresponding number of idle resources for each mirror image according to the number of the idle resources allocated to each hot mirror image.
The hot images counted by the hot image counting module are not always allocated to idle resources.
This depends on the ranking of the hot images, and when the number of idle resources of the hot image with the top ranking is less than 1 and 1 is taken, the number of idle resources of the hot image with the bottom ranking is occupied, so that the probability that the number of idle resources cannot be allocated is higher the later the ranking is.
The number of free resources allocated to the last hot image of the name may be greater than the number of resources allocated according to the weight. When the number of idle resources of the hotspot image with the top rank is a non-integer greater than 1, because of taking an integer, the discarded decimal part may increase the number of remaining idle resources, which may result in that the number of idle resources allocated to the last hotspot image is greater than the number of resources allocated according to the weight.
Compared with the existing method for directly creating the practical training cloud environment, the method can intelligently and automatically count the experimental environment with higher use frequency and create enough experimental environment in advance to be placed in the resource pool, and can automatically provide an idle experimental environment from the resource pool to the practical training cloud when in need.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, and it should be understood that the above-mentioned embodiments are only examples of the present invention and are not intended to limit the scope of the present invention. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the invention, may occur to those skilled in the art and are intended to be included within the scope of the invention.
Claims (10)
1. A resource pool system for rapidly providing an experimental environment, comprising: the system comprises a resource pool configuration module, a heat mirror management module, a resource scheduling module, a resource counting module and a cloud driving module;
the resource pool configuration module is used for configuring a resource pool system and controlling the resource pool to be started and closed; the configuration of the resource pool comprises the total size setting of the resource pool, the minimum idle resource number setting of the resource pool, the hot spot mirror image statistical method setting and the resource scheduling method setting;
the heat mirror image management module is used for counting the heat mirror images according to a configured heat mirror image counting method;
the resource scheduling module is used for performing dynamic telescopic allocation on the resources for the counted hotspot mirror images according to the configured resource scheduling method and the minimum idle resource number of the resource pool;
the resource counting module is used for counting resource data of each hotspot mirror image of the resource pool;
the cloud driving module is used for butting a cloud computing platform supported by the bottom layer and calling an interface of the cloud computing platform.
2. The resource pool system for rapidly providing experimental environments as claimed in claim 1, wherein the heat image management module performs statistics on heat images through a full image statistical method, and the full image statistical method specifically includes:
counting the previous n mirror image historical use records;
sorting the n records according to mirror names, installing the records, and sequencing the records into X according to the use times;
classifying the X according to the same use times of the mirror images and sequencing the X into Y according to the times;
sorting the n records according to the mirror image service time and removing the duplicate into Z;
allocating the name times to the corresponding mirror image of the Y according to the times from large to small;
and recording the weight of the corresponding mirror image, wherein the using times of the mirror image is the weight of the mirror image.
3. The resource pool system for rapidly providing experimental environments of claim 2, further comprising, prior to the recording the weights of the corresponding images: when the use times of a plurality of images are the same, ranking is carried out according to the order of the images in Z.
4. The resource pool system for rapidly providing experimental environments as claimed in claim 1, wherein the heat image management module performs statistical heat image by a quantitative image statistical method, and the quantitative image statistical method specifically comprises:
counting the previous n mirror image historical use records;
sorting the n records according to mirror names, installing the records, and sequencing the records into X according to the use times;
classifying the X according to the same use times of the mirror images and sequencing the X into Y according to the times;
allocating the name times to the corresponding mirror image of the Y according to the times from large to small;
and recording the weight of the corresponding mirror image, wherein the using times of the mirror image is the weight of the mirror image.
5. The resource pool system for rapidly providing experimental environments of claim 4, further comprising, prior to the recording the weights of the corresponding images: when the use times of a plurality of mirror images are the same, counting the top 2n mirror images in the historical use record, sequencing the plurality of mirror images according to the times and distributing the names.
6. The resource pool system for rapidly providing experimental environments as claimed in claim 1, wherein the resource scheduling module is configured to perform the step of dynamically scaling and allocating resources, and specifically includes:
acquiring a configured resource scheduling mode, and executing a scheduling task according to the resource scheduling mode;
acquiring a hot mirror noun and weight, and acquiring the minimum idle number of a resource pool;
performing heat mirror image idle resource number distribution according to the ranking, and taking 1 when the distributed number is less than 1; when the distributed number is an integer, taking the integer; when the distributed number is a non-integer greater than 1, taking an integer part; and the cloud driving module creates idle resources for each mirror image according to the number of the idle resources allocated to each mirror image.
7. The resource pool system for rapidly providing experimental environments of claim 6, wherein the resource scheduling manner is a timing scheduling manner, and the timing scheduling manner comprises: and setting a scheduling time point, and triggering a resource scheduling task when the system time reaches the scheduling time point.
8. The resource pool system for rapidly providing experimental environments of claim 6, wherein the resource scheduling manner is a real-time scheduling manner, and the real-time scheduling manner comprises: and monitoring the change of the hot mirror image in real time, and triggering a resource scheduling task when the judgment is changed.
9. The resource pool system for rapidly providing experimental environments of claim 6, wherein the resource scheduling manner is a manual scheduling manner, and the manual scheduling manner comprises: and directly triggering the resource scheduling task.
10. The resource pool system for rapidly providing experimental environments of claim 6, wherein the resource scheduling manner is a low load scheduling manner, and the low load scheduling manner comprises: and monitoring the load condition of the cloud computing platform in real time, and directly triggering the resource scheduling task when the load of the cloud computing platform is detected and judged to be lower than a set threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911221886.1A CN111078402A (en) | 2019-12-03 | 2019-12-03 | Resource pool system capable of rapidly providing experimental environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911221886.1A CN111078402A (en) | 2019-12-03 | 2019-12-03 | Resource pool system capable of rapidly providing experimental environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111078402A true CN111078402A (en) | 2020-04-28 |
Family
ID=70312613
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911221886.1A Pending CN111078402A (en) | 2019-12-03 | 2019-12-03 | Resource pool system capable of rapidly providing experimental environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111078402A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521055A (en) * | 2011-12-15 | 2012-06-27 | 广州杰赛科技股份有限公司 | Virtual machine resource allocating method and virtual machine resource allocating system |
| CN104216784A (en) * | 2014-08-25 | 2014-12-17 | 杭州华为数字技术有限公司 | Hotspot balance control method and related device |
| CN104331328A (en) * | 2013-07-22 | 2015-02-04 | 中国电信股份有限公司 | Virtual resource scheduling method and virtual resource scheduling device |
| CN105511959A (en) * | 2014-10-16 | 2016-04-20 | 腾讯科技（深圳）有限公司 | Method and device for distributing virtual resource |
| CN106302632A (en) * | 2016-07-21 | 2017-01-04 | 华为技术有限公司 | The method for down loading of a kind of foundation image and management node |
| CN106528262A (en) * | 2015-09-10 | 2017-03-22 | 华为技术有限公司 | Image deployment method and device |
-
2019
- 2019-12-03 CN CN201911221886.1A patent/CN111078402A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521055A (en) * | 2011-12-15 | 2012-06-27 | 广州杰赛科技股份有限公司 | Virtual machine resource allocating method and virtual machine resource allocating system |
| CN104331328A (en) * | 2013-07-22 | 2015-02-04 | 中国电信股份有限公司 | Virtual resource scheduling method and virtual resource scheduling device |
| CN104216784A (en) * | 2014-08-25 | 2014-12-17 | 杭州华为数字技术有限公司 | Hotspot balance control method and related device |
| CN105511959A (en) * | 2014-10-16 | 2016-04-20 | 腾讯科技（深圳）有限公司 | Method and device for distributing virtual resource |
| CN106528262A (en) * | 2015-09-10 | 2017-03-22 | 华为技术有限公司 | Image deployment method and device |
| CN106302632A (en) * | 2016-07-21 | 2017-01-04 | 华为技术有限公司 | The method for down loading of a kind of foundation image and management node |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111680124B (en) | RPA-based large-scale custom customer demand acquisition and query method | |
| US9996593B1 (en) | Parallel processing framework | |
| CN108920153A (en) | A kind of Docker container dynamic dispatching method based on load estimation | |
| CN107102788A (en) | Application notification | |
| WO2019062699A1 (en) | Resource scheduling method, scheduling server, cloud computing system and storage medium | |
| CN110071965B (en) | Data center management system based on cloud platform | |
| CN108509280B (en) | Distributed computing cluster locality scheduling method based on push model | |
| CN102541467A (en) | Scheduling method for read-write requests of object data server | |
| CN105530169A (en) | Group creating method and device | |
| WO2022257302A1 (en) | Method, apparatus and system for creating training task of ai training platform, and medium | |
| CN108959399A (en) | Distributed data deletes flow control method, device, electronic equipment and storage medium | |
| CN105022668B (en) | Job scheduling method and system | |
| CN103491024A (en) | Job scheduling method and device for streaming data | |
| WO2013139037A1 (en) | Method and device for scheduling resources | |
| CN107273200A (en) | A kind of method for scheduling task stored for isomery | |
| CN110750331A (en) | Container cluster scheduling method and platform for education desktop cloud application | |
| US10592107B2 (en) | Virtual machine storage management queue | |
| US20180081894A1 (en) | Method and apparatus for clearing data in cloud storage system | |
| CN113296874A (en) | Task scheduling method, computing device and storage medium | |
| US10402370B2 (en) | Information processing method and electronic apparatus | |
| CN105550025B (en) | Distributed infrastructure services (IaaS) dispatching method and system | |
| CN105487946B (en) | A kind of faulty computer automatic switching method and device | |
| CN108696373A (en) | Virtual resource allocation method, NFVO and system | |
| US11488115B1 (en) | Efficient meeting room reservation and scheduling | |
| CN111405072B (en) | Hybrid cloud optimization method based on cloud manufacturer cost scheduling |
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 |