JPH0358222B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] The system classifies calls made by an exchange into multiple call types, determines the number of calls that can be accepted for each call type, and allows each call type to be independently issued for a certain period of time. The number of calls is registered, and when the number of calls exceeds the allowable number of calls determined for each call type, calls to the corresponding call type are suppressed.

[Industrial application field]

The present invention relates to a dynamic overload control method in an exchange.

In a stored program control system without an overload control function, when overloaded, the central processing unit will receive all incoming and outgoing calls that exceed its processing capacity, resulting in an increase in waiting queues in the system, waiting for resource acquisition, etc. As a result, the throughput in the system becomes lower than normal. An overload control mechanism is required to prevent this, but
Considering that the processing mechanism of the stored program control method works safely even when the central processing unit usage rate is relatively high, it exceeds the highest central processing unit (hereinafter abbreviated as CC) usage rate that is not adversely affected by overhead. It is desirable to strike a balance between allowing and restricting incoming and outgoing calls so as to provide maximum serviceability to subscribers.

The system of the present invention described below is one that reflects this idea.

[Conventional technology]

A conventional overload control method will be explained with reference to FIG. Figure 8 shows the CC usage rate on the vertical axis and time on the horizontal axis.The regulation method shown here uses the regulation issuing value (ηr: expressed as CC usage rate) as a parameter for regulation processing. ) and a deregulation value (ηc). And for example in a 20 second period
The CC usage rate (η) is measured, and restrictions are issued when η≧ηr. During regulation, if η≧ηc, the regulation level is strengthened, and when η<ηc, it is relaxed.

[Problem that the invention seeks to solve]

In the method shown in Figure 8, ηr > ηc, so CC
Usage rate increases with time when regulation does not work
It's fluctuating under. However, once ηr is exceeded and restrictions are imposed, restrictions will continue until the CC usage rate falls to the restriction release value. During this period, the exchange is under restrictions and calls from the exchange are restricted. In other words, even if the regulation is no longer necessary, the regulation continues until the release value, resulting in excessive regulation.

The present invention has been developed in view of these drawbacks, and by detecting the allowable number of accepted calls at appropriate intervals, dynamic overload can be realized that accurately corresponds to changes in call types and overload conditions at each point in time. The purpose is to provide a load control method.

[Means for solving problems]

FIG. 1 is a block diagram showing the principle of the system of the present invention.

In FIG. 1, block 101 is a CC overload monitoring processing section, block 102 is a processing section for determining the number of allowable calls to be accepted, block 103 is an outgoing call processing section,
Block 104 is an incoming call processing unit, and block A is a CC.
Overload display memory, block B is average
CC usage rate display memory, block C indicates overload activation value memory, block D indicates CC overload release value memory, block E indicates memory for accumulating the number of calls accepted in the previous cycle, and block F indicates memory for setting the number of calls allowed to be accepted in the next cycle. .

The CC overload monitoring processing section of block 101 checks whether there is an overload display at a certain period (for example, 20 seconds) (block A),
It monitors what the average CC usage rate is (block B), whether it is necessary to activate an overload (block C), and whether it is necessary to release the overload in the case of an overload (block D).

The block 102, which determines the number of calls allowed to be accepted, calculates the contents of the average CC usage rate display memory and the contents of the memory for accumulating the number of calls accepted in the previous cycle in a short cycle (for example, 4 seconds) in which the regulation can follow fluctuations in calls. Set the allowable number of calls to be accepted in the next cycle and save it to memory F.
write to.

The outgoing call processing section 103 and the incoming call processing section 104 perform call processing based on the contents of the next cycle acceptance allowable call number setting memory F, which is set based on the contents of the previous cycle acceptance call number accumulation memory E.

[Effect]

Find the CC usage rate and the total number of incoming and outgoing calls up to the previous cycle for each call type, and calculate how many incoming and outgoing calls can be accepted at the maximum CC usage rate when the system operates with the same call type ratio as in the previous cycle. If there are incoming or outgoing call requests that exceed these limits, restrictions will be placed on each type of call. By setting this allowable number of acceptances at appropriate intervals, it becomes possible to carry out regulation control that appropriately responds to changes in call types and overload conditions at each point in time.

FIG. 2 shows an example of an overload control curve, that is, a CC usage rate transition curve obtained by the method according to the present invention.

Figure 2 shows that after overload detection, restrictions are applied to each call type, and unlike the conventional system, the CC usage rate is controlled in stages, which prevents sudden restrictions in switching connection processing. It shows.

〔Example〕

Embodiments of the system according to the present invention will be described in detail below with reference to the drawings.

First, the algorithm for determining the number of calls allowed to be accepted in the next cycle is determined as follows.

(Number of calls allowed to be accepted in the next cycle) = [Maximum allowable value of CC usage rate] - [CC usage rate at no load] / [average CC usage rate in the previous cycle)] - [CC usage rate at no load]
×[Number of calls accepted in the previous cycle] In order to prevent extreme increases and decreases in traffic, the following correction formula is used. In other words, (a) (Number of calls accepted in the next cycle) ≧ (Number of calls accepted in the previous cycle) x (1 + maximum allowable increase rate) (Number of calls accepted in the next cycle) = (Number of calls accepted in the previous cycle) x ( 1 + Maximum allowable increase rate) (b) (Number of calls accepted in the next cycle) ≦ (Number of calls accepted in the previous cycle) × (1 - Minimum allowable increase rate) (Number of calls accepted in the next cycle) = (Number of calls accepted in the previous cycle) Number of accepted calls) x (1 - minimum allowable increase rate) The procedure for implementing this algorithm will be described in detail below.

FIG. 3A is a flowchart showing the operation of the CC overload monitoring processing section.

Block 301 monitors CC overload for calls with a period of 20 seconds, for example. In block 302, it is determined from the contents of the CC overload display memory A whether or not CC overload is in progress. If it is determined in block 302 that the device is not overloaded, block 3
03 Average CC usage rate display Average of memory B
Determine whether the CC usage rate is greater than the overload activation value α, and if it is greater or not equal, terminate the process, and if greater or equal, block 3
04, the CC overload display memory A is set to "1" and the program for determining the number of calls that can be accepted is activated. If it is determined in block 302 that the CC overload is being overloaded from the contents of the CC overload display memory A, the process proceeds to block 305;
It is determined whether the average CC usage rate is smaller than the CC overload release value β (block D), and if it is larger, the process is terminated, and if it is smaller, it is moved to block 306 and the CC overload display memory is Write "0" to , cancel the overload control process, and stop the program for determining the number of calls that can be accepted.

FIG. 4 is a flowchart illustrating the procedure for determining the number of calls allowed to be accepted. In block 401, the number of calls to be accepted in the next cycle is determined, for example, every four seconds. That is, in block 402, the number of calls allowed to be accepted in the next cycle is calculated from the memory E for accumulating the number of calls accepted in the previous cycle and the memory B for displaying the average CC usage rate, using the above-mentioned formula, and written into the memory F for the number of calls allowed to be accepted in the next cycle. After finishing block 4
03 clears the memory E for accumulating the number of calls accepted in the previous cycle.

FIG. 5 is a flowchart showing the procedure for processing an outgoing call. Block 501 initiates processing each time an outgoing call occurs. block 502
In block 503, the contents of the CC overload display memory A are used to determine whether overload is occurring, and in block 503, the number of accepted calls is registered based on the contents of the outgoing call in the memory for the number of accepted calls in the previous cycle. Next, in block 504, the number of calls allowed to be accepted is subtracted, and the number of calls allowed to be accepted in the next cycle is written in the outgoing call column of the memory F for the number of calls allowed to be accepted in the next cycle. Next, in block 505, it is determined whether or not the outgoing call can be accepted. If it is possible, the process goes to acceptance processing, and if it is not possible, it goes to restriction processing.

FIG. 6 is a flowchart showing the procedure of incoming call processing. In block 601, processing is started every time an incoming call occurs. In block 602, it is read whether or not overload is occurring based on the contents of the CC overload display memory A, and in block 603, the number of accepted calls is registered based on the contents of the incoming calls in the memory for the number of accepted calls in the previous cycle. Next, in block 604, the number of calls allowed to be accepted is subtracted, and the number of calls allowed to be accepted in the next cycle is written in the incoming call column of the memory F for the number of calls allowed to be accepted in the next cycle. Then block 6
In step 05, it is determined whether or not the incoming call can be accepted, and if it is possible, the process goes to reception processing, and if it is not possible, it goes to regulation processing.

FIG. 7 shows the system configuration of the system according to the present invention, in which 701a and 701b are multiple networks, 702a and 702b are central processing units (call processing processors), 703a and 703b are main memories, and 704 is a central Processing device (main processor), 705 is main memory, 706a, 706b are interoffice lines (output/input trunk), 707a, 707
b represents a subscriber line concentrator, and 708a and 708b represent subscribers, respectively.

〔Effect of the invention〕

As explained in detail above, according to the present invention, by updating the allowable number of accepted calls at appropriate intervals, it is possible to carry out regulatory control that accurately responds to changes in call types and overload conditions at each point in time. , it is possible to eliminate excessive regulation as in the past, and the serviceability of exchange connections can also be improved.

[Brief explanation of drawings]

Fig. 1 is a flowchart for explaining the principle of the method according to the present invention, Fig. 2 is a transition diagram of the CC usage rate in the method according to the present invention, and Fig. 3 A and B are the methods according to the present invention. Fig. 4 is a flowchart showing the procedure for determining the number of calls allowed to be accepted, Fig. 5 is a flowchart showing the procedure for outgoing call processing, and Fig. 6 is a flowchart showing the procedure for processing incoming calls. 7 is a system configuration diagram of the method according to the present invention, and FIG. 8 is a transition diagram of the CC usage rate in the conventional method. In FIG. 1, block 101 is a CC overload monitoring processing section, block 102 is a processing section for determining the number of allowable calls to be accepted, block 103 is an outgoing call processing section,
Block 104 is an incoming call processing unit, and block A is a CC.
Overload display memory, block B is average
Block C shows the memory for displaying the CC usage rate, block C shows the overload activation value memory, block D shows the CC overload release value, block E shows the memory for accumulating the number of calls accepted in the previous cycle, and block F shows the memory for setting the number of calls allowed to be accepted in the next cycle.

Claims (1)

[Scope of Claims] 1. A monitoring processing unit 101 that monitors the overload of the central processing unit of the exchange, and updates the number of calls that can be accepted at appropriate intervals, and determines the number of calls that can be accepted for each type of call at each predetermined period. Processing part 1
02, which classifies calls made when the central processing unit is overloaded into multiple call types, determines the allowable number of calls for each call type, and independently calculates the number of calls made for each call type. A dynamic overload control method for an exchange, which is characterized by counting the number of calls, and suppressing calls to the corresponding call type if the number of calls exceeds the permissible number for each call type. 2. The allowable number of calls for each of the call types is determined based on the usage rate of the exchange and the central processing unit and the number of calls accepted for each call type. Dynamic overload control method for switching equipment. 3 The allowable number of calls for each of the above call types is determined based on the usage rate of the central processing unit of the exchange and the total number of calls received for each call type. 2. A dynamic overload control system for an exchange according to claim 1, wherein the dynamic overload control method for an exchange is determined by distribution determined by the number of calls received. 4. The allowable number of calls for each of the above call types is determined based on the usage rate of the central processing unit of the exchange and the total number of calls received for each call type. 2. A dynamic overload control system for an exchange according to claim 1, wherein the dynamic overload control method for an exchange is determined by distribution determined by the number of calls made. 5 The allowable number of calls for each of the above call types is determined based on the usage rate of the central processing unit of the exchange and the total number of calls received for each call type. 2. A dynamic overload control system for an exchange according to claim 1, wherein the dynamic overload control system is determined by distribution determined by the number of calls suppressed.