CN1235419C - Method for resource planning method in communication system - Google Patents

Abstract

The invention discloses a resource planning method in a communication system. In the method, part of system resources are set as reserved resources reserved for high-priority users, and other system resources are shared by high-priority users and low-priority users. resources, when a user applies for network access, first allocate resources to the user from the shared resources and provide services for the user, if the shared resources have been allocated, then determine whether the user is a high-priority user, if so, from the reserved resources Allocate resources for the high-priority user and provide services for the user. When the shared resources have been allocated and there is no resource that can be allocated to the high-priority user in the reserved resources, select a low-priority user for resource preemption. Use preemption resources serve high-priority users; the above scheme can avoid or reduce resource preemption to some extent; at the same time, it effectively solves the problem that the system provides priority services for high-priority users.

Description

A Resource Planning Method in Communication System

technical field

The invention relates to a resource allocation and management method in a communication system.

Background technique

In current communication systems, especially in various mobile communication systems, traffic peaks may occur in some cases, and at this time, the total amount of resources requested by users is greater than the total amount of available resources in the system. The occurrence of this situation will cause some users to be unable to access the system, or cannot obtain satisfactory services after accessing the system, that is, the quality of service (QOS) of their traffic applications cannot be satisfied. Generally, for network operators, when there is a traffic peak, it is often necessary to fully or as far as possible guarantee the network access of a specific user in its user group or meet the service quality requirements of the network traffic already connected. Therefore, the above requirements need to be realized through the planning and management of system resources. The most common way to meet the above requirements is to reject the access requests of some general users or interrupt their services or reduce the quality of service provided for general users. For example, the most common way to achieve the above purpose is to use the system resource preemption method, which is realized by monitoring the use of system resources. When it is found that the system resources are insufficient, the services of users with lower priority levels are forcibly interrupted to release resources for use by users with higher priority levels. This method is reasonable from the perspective of resource utilization efficiency, and it can manage and use system resources according to the principle of the highest benefit of resource utilization. However, in actual network operation, interrupting a user's business is more unacceptable than delaying or waiting for network access, no matter for the operator or the user. Therefore, the system always tries to avoid the use of this method in practice.

As an alternative to the above method, a method based on priority queuing may also be used to solve the problem of system resource usage. The essence of this method is: when the system resources are insufficient, the applicants wait in line, and the specific queuing is completed in order according to the user's priority. The business of users with high priority is always ranked first in the queue, and users with the same priority are in order according to the application time. Queuing, so that once the system resources are free, high-priority users will be served first.

Among the above two methods, the biggest harm of the first method is the reduction of service reputation, and the side effects are relatively large; for the second method, the queuing waiting time that users can tolerate when accessing is usually shorter, especially When applying for voice services, the general choice is to re-initiate the call, and in such a short period of time, the probability of obtaining resources due to the end of other users' traffic is not high, so this solution is not very effective in practical applications .

Contents of the invention

The purpose of the present invention is to provide a method for resource planning in a communication system, which can effectively solve the problem that the system provides priority services for high-priority users, and at the same time avoid or reduce the occurrence of resource preemption.

In order to achieve the above purpose, the resource planning method in the communication system provided by the present invention includes:

Set the amount of system resources reserved for high-priority users, and use the remaining system resources as shared resources for high-priority users and low-priority users;

When a user applies for network access, first allocate resources for the user from the shared resources and provide services for the user. If the shared resources have been allocated, it is judged whether the user is a high-priority user. If so, the resource is reserved for the user. High-priority users allocate resources and provide services for users, otherwise the low-priority users are denied network access.

When the shared resources have been allocated and there is no resource available for high-priority users in the reserved resources, low-priority users are selected for resource preemption, and the preempted resources are used to serve high-priority users.

Before selecting low-priority users for resource preemption, send high-priority users who apply for network access to the high-priority user queue for queuing. When the queuing time exceeds the specified threshold, select low-priority users for resource preemption. seize.

When the queuing time of a high-priority user does not exceed the specified threshold, and the system already has shared resources or reserved resources that can be allocated to the high-priority user, it will allocate resources to the high-priority user and provide services for it, and clear The queuing timer for this high-priority user.

Before rejecting the low-priority user's network access, send the low-priority user to the low-priority user queue for queuing, and when the queuing time exceeds the specified threshold, then reject the low-priority user's network access.

In addition, the present invention can determine the system resources reserved for high-priority users according to the ratio of the number of high-priority users and the number of low-priority users in the system; it can also dynamically adjust the system resources reserved for high-priority users according to the historical occupancy information of system resources. The number of system resources used by priority users; if the latter is used, the following method is used to dynamically adjust the number of system resources reserved for high-priority users: If the behavior of preempting shared resources occurs within the specified time before the current time , then increase the amount of system resources reserved for high-priority users. If low-priority users fail to access the network or queue up within the specified time before the current time, and the reserved resources are not used or the usage rate is low If the value is lower than the specified threshold, the amount of system resources reserved for high-priority users is reduced.

When increasing the number of system resources reserved for high-priority users, first check whether there are idle resources in the shared resources. If so, check whether the low-priority user queue is empty. If it is empty, set the idle shared resources Reserve resources.

The present invention also sets a threshold for increasing the number of reserved resources, and sets within the range of the threshold when setting idle shared resources as reserved resources.

Since the present invention adopts the method of setting reserved system resources for high-priority users, and uses the remaining system resources as shared resources for high-priority users and low-priority users, in this way, when users apply for network access, they can firstly start from the shared resources. Resources are allocated to users and services are provided to users. If shared resources have been allocated and the user is a high-priority user, resources can be allocated to the high-priority user from the reserved resources and services are provided to the user. , which can avoid or reduce the occurrence of resource preemption to some extent; at the same time, when the shared resources have been allocated and there are no resources that can be allocated to high-priority users in the reserved resources, then select low-priority users for Resource preemption, using preempted resources to serve high-priority users, thus effectively solving the problem that the system provides priority services for high-priority users.

Description of drawings

Fig. 1 is the flow chart of the embodiment of the method of the present invention;

Fig. 2 is a flow chart of adaptively adjusting reserved resources adopted in the embodiment described in Fig. 1;

FIG. 3 is a flow chart of adding reserved resources used in the process shown in FIG. 2 .

Detailed ways

In the actual communication system, it is inevitable that high-priority users preempt the system resources occupied by low-priority users when there is a traffic peak. It is even more unacceptable, so it is necessary to avoid this phenomenon as much as possible in practice. Based on this, the essence of the present invention is to reserve a certain amount of system resources for high-priority users through the planning and management of system resources, and use the remaining resources as shared resources for high- and low-priority users to ensure When resources are insufficient, reserve resources to provide services for high-priority users, try to avoid preempting resources for low-priority users, and ensure the quality of services for high-priority users.

The specific implementation scheme of the present invention is as follows: firstly, the amount of system resources reserved for high-priority users is set, and the rest of the system resources are used as shared resources for high-priority users and low-priority users. The specific reservation value can be configured through the system network management center. At the beginning, the number of reserved resources can be set according to experience, or according to the number of high-priority users and the number of low-priority users contracted with the network operator. to determine the ratio. For example, 10% of reserved resources can be set for high-priority users, and the remaining 90% of resources are shared resources, shared by high-priority and low-priority users. In this way, under normal circumstances, when a user applies for network access, the user is first allocated resources from the shared resources and provides services for the user, that is, only resources are allocated in the shared resource pool under normal circumstances, but if the shared resources have been allocated , for new resource applications due to new users accessing the system, it is necessary to determine whether the user is a high-priority user, and if so, allocate resources for the high-priority user from the reserved resources and provide services for the user , if it is a low-priority user, the network access of the low-priority user. However, in practice, there is still such a situation, that is, when high-priority users access, shared resources have been allocated, and there are no resources that can be allocated for high-priority users in the reserved resources. At this time, in order to ensure that the system is high-priority The quality of service provided by users, select low-priority users for resource preemption, and then use the preempted resources to serve high-priority users.

In the above method, in order to further reduce the number of preemption of resources occupied by low-priority users, when high-priority users apply for resources and there are no idle resources in the reserved resources, it can be considered to let them join the priority queue first. When its queuing time exceeds a certain threshold (for example, 5 seconds) and still has not waited for the resource allocated to it, it starts to implement resource preemption for low-end low-priority users. Therefore, it is also necessary to pre-set the queuing queue of high-priority users and the queuing threshold of high-priority users. In this way, when high-priority users access the system and the system has no resources that can be allocated to high-priority users, Then, before selecting low-priority users for resource preemption, send high-priority users who apply for network access to the high-priority user queue for queuing, and start the queuing timer of the high-priority user at the same time. When it is known that the queuing time exceeds the specified threshold, users with lower priority levels are selected to preempt resources. If the queuing time of the high-priority user does not exceed the specified threshold and the system already has shared resources or reserved resources that can be allocated to the high-priority user, then allocate resources to the high-priority user and provide services for it, At the same time, clear the queuing timer of the high-priority user.

Similarly, in order to make the system function more perfect, if there is no resource to allocate when the low-priority user accesses the network, it can also be considered to allow it to participate in the priority queue instead of rejecting it immediately. Therefore, in order to realize the above operations, it is also necessary to pre-set the queuing queue of low-end low-priority users and the queuing threshold of low-end low-priority users. In this way, before denying the network access of low-priority users, the Low-priority users are sent to the low-priority user queue for queuing, and the queuing timer of the low-priority user is started at the same time. When the queuing time exceeds the specified threshold, the network access of the low-priority user is rejected. Of course, if the system already has shared resources that can be allocated to the low-priority user when the queuing time of the low-priority user does not exceed the specified threshold, the system will allocate resources to the low-priority user and provide services for it, and clear The queuing timer for this low-priority user.

In the specific implementation process of the present invention, since the occupancy or usage of system resources is changing dynamically, the predetermined value is only determined according to experience or the ratio of the number of high-priority users and the number of low-priority users contracted with the network operator. The amount of resources reserved for high-priority users may be unreasonable, because such statically set values are usually difficult to fully reflect changes in system resource usage, and may also lead to underutilization of system resources. Therefore, the present invention also dynamically adjusts the amount of system resources reserved for high-priority users according to the historical occupancy information of system resources, that is, determines a reliable traffic model according to historical statistical information, and on this basis, provides high-priority users The number of reserved resources is dynamically adjusted according to the resource usage in the previous period. The specific dynamic adjustment method is as follows: If the behavior of preempting shared resources occurs within the specified time before the current time, it indicates that the resource reservation is small, and you can consider increasing the resources reserved for high-priority users in the next time period The number of system resources, for example, the released shared area resources are immediately designated as reserved resources; if low-priority user network access failure or queuing behavior occurs within the specified time before the current time, and the reserved resources are not used or used If the rate is lower than the specified threshold, it indicates that the amount of reserved resources is too large, and you can consider reducing the amount of system resources reserved for high-priority users.

In the specific process of increasing the number of system resources reserved for high-priority users, first check whether there are idle resources in the shared resources. If there are, it means that there are already reserved resources that can be increased. Check whether the level user queue is empty, if it is empty, it means that the shared resources have resources that can be reserved, so set the idle shared resources as reserved resources. In order to minimize the impact of the adjustment of the number of reserved resources on the system, in practice, a threshold for increasing the number of reserved resources can be set, and when idle shared resources are set as reserved resources, set or increase within the range of the threshold.

Similarly, in the specific process of reducing the number of system resources reserved for high-priority users, first check whether there are idle resources in the reserved resources. If there are, it means that there are already reserved resources available for reduction. At this time, Then check whether the high-priority user queue is empty. If it is empty, it means that the shared resource has resources that can be shared. Therefore, set the idle reserved resource as a shared resource. Of course, when reducing reserved resources, you can also set a threshold for the reduced number to reduce the impact on the system when dynamically adjusting the number of reserved resources.

In addition, in order to make the resource reservation more accurate, a shorter time period may be considered, for example, the resource reservation amount may be adjusted every 30 minutes.

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Fig. 1 is a flowchart of an embodiment of the method of the present invention. According to Figure 1, when a user applies for system resources through network access in step 1, the system checks in step 2 whether there are idle resources in the shared resources. Provide services, if there is no free resource, then judge whether the user is a high-priority user in step 4, if not, reject the user's network access request or do other processing in step 5; if it is a high-priority user, then Continue to judge whether there are idle resources in the reserved resources, if yes, allocate resources for the user from the shared resources in step 7, and provide services for the user; if there is no resource that can be allocated for the user in the shared resources, then in In step 8, the user is sent to the preset high-priority user queue for queuing, and a timer is set for the user, and finally in step 9, corresponding processing is performed on the user according to the state of the timer. The specific operation of step 9 is as follows: before the timer expires, if the user voluntarily withdraws from the network access, then clear the user's timer; if the user waits for idle resources before the timer expires, the user Allocate resources, provide services for the user, and clear the timer; if the timer expires, select a low-priority user from the shared resource to preempt the resource.

During the running process of the embodiment shown in FIG. 1 , the amount of resources reserved for high-priority users needs to be adaptively adjusted. Refer to FIG. 2 for the specific process. First set a T minute timer in step 21, and then check various statistical values related to system resource usage in step 22, such as the number of high-priority users and low-priority users currently connected to the network, reserved resources and shared Resource utilization etc., then judge in step 23 whether the resource preemption to the low-priority user has occurred in the past T minutes, if resource preemption phenomenon has occurred, it shows that reserved resources are insufficient, then a mark J is set in step 25 , to indicate that K reserved resources will be added within the next T minutes, and the K value can be set according to experience values, and can also be dynamically adjusted according to the utilization of system resources. If it is determined in step 23 that no resource preemption has occurred in the past T minutes, it means that the number of reserved resources does not need to be increased or adjusted at present, but whether it needs to be decreased or adjusted needs to be further judged. Therefore, continue to judge in step 24 whether the phenomenon that the low-priority user was denied network access has occurred in the past T minutes. Instead, continue to perform other system operations; if low-priority users are denied access to the network in the past T minutes, it means that the shared resources are insufficient. Part of the reserved resources is converted into shared resources, and it is necessary to continue to judge through step 27 whether the maximum utilization rate of reserved resources in the past T minutes is less than the preset threshold. If it is less than, it means that the reserved resources are too large and can be converted Some of them are used as shared resources, so in step 28, convert some of the reserved resources into shared resources according to the preset conversion value; otherwise, the amount of reserved resources is appropriate and cannot be converted, so no resource adjustment is performed in step 29 operation, but continue with other system operations.

In the adjustment process of the above-mentioned reserved resource resources, the increase of K reserved resources described in step 25 is implemented according to the following process, referring to Fig. Sort out, then check in step 252 whether the low-priority user queue has queued users, if there is, then in step 254 allocate resources for the queued users, and provide services to it, then proceed to step 257; if check low-priority in step 252 Level user queue does not have the user of queuing, then in step 253, check whether the system is set to expand the mark J of reserved resource, if not, explain and do not need to adjust reserved resource, end adjustment process in step 256, otherwise this idle in step 255 The shared resources are converted into reserved resources according to the preset strategy; then in step 257, check whether the adjustment amount of the reserved resources reaches the requirement of the preset K value, if so, end the adjustment process in step 259, and clear the increase The mark J of reserved resource; Otherwise also will go through step 258 and judge whether T minute timer is overtime, if overtime, also execute step 259, if not overtime, then return to step 251 and continue to increase the adjustment of reserved resources.

In a word, through the resource planning method described in the present invention, the phenomenon of resource preemption can be avoided or reduced to the greatest extent, and the quality of service provided for high-priority users can be guaranteed.

Claims (10)

1. A resource planning method in a communication system, comprising: Set the amount of system resources reserved for high-priority users, and use the remaining system resources as shared resources for high-priority users and low-priority users; When a user applies for network access, first allocate resources for the user from the shared resources and provide services for the user. If the shared resources have been allocated, it is judged whether the user is a high-priority user. If so, the resource is reserved for the user. High-priority users allocate resources and provide services for users, otherwise the low-priority users are denied network access. 2. The resource planning method according to claim 1, wherein when the shared resources have been allocated and there is no resource that can be allocated to high-priority users in the reserved resources, select low-priority users to preempt resources , using preempted resources to serve high-priority users. 3. The resource planning method according to claim 2, wherein before selecting low-priority users for resource preemption, the high-priority users applying for network access are sent to the high-priority user queue for queuing. When the queuing time exceeds the specified threshold, users with lower priority levels are selected for resource preemption. 4. The resource planning method according to claim 3, wherein when the queuing time of high-priority users does not exceed the specified threshold, and the system already has shared resources or reserved resources that can be allocated to high-priority users , allocate resources for the high-priority user and provide services for it, and clear the queuing timer of the high-priority user at the same time. 5. The resource planning method according to claim 1, characterized in that before rejecting the network access of the low-priority user, the low-priority user is sent to the low-priority user queue for queuing, and when the queuing time exceeds the specified When the threshold is reached, the network access of the low-priority user is denied. 6. The resource planning method according to claim 1, 2, 3, 4 or 5, characterized in that, according to the ratio of the number of high-priority users and the number of low-priority users in the system, it is determined to reserve for high-priority users System resources used. 7. The resource planning method according to claim 6, wherein the amount of system resources reserved for high-priority users is dynamically adjusted according to historical occupancy information of system resources. 8. The resource planning method according to claim 7, wherein the following method is used to dynamically adjust the amount of system resources reserved for high-priority users: if a shared resource is preempted within a specified time before the current time increase the amount of system resources reserved for high-priority users. If low-priority users fail to access the network or queue up within the specified time before the current time, and the reserved resources are not used or used rate falls below the specified threshold, the amount of system resources reserved for high-priority users is reduced. 9. The resource planning method according to claim 8, wherein when increasing the number of system resources reserved for high-priority users, first check whether there are idle resources in the shared resources, and then check the low-priority Whether the level user queue is empty, if it is empty, set the idle shared resources as reserved resources. 10. The resource planning method according to claim 9, characterized in that a threshold for increasing the number of reserved resources is set, and when idle shared resources are set as reserved resources, it is set within the range of the threshold.

Images