JP2010016459A - Request reception system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a request reception system which correctly obtains an operation state, carries out feedback operation, and can optimize access to an end server without modification of the end server. <P>SOLUTION: The system includes a correction server HS which receives a service request of a sample terminal MN(S) selected out of user terminals MN, transfers to the end server ES, receives service offered from the end server ES, and transfers to the user terminal MN. The correction server HS includes: a service time-measuring unit 36 which measures time from start to the end of service which the end server ES answers the service request and offers the service; and a reporting unit 38 which reports the service time to an access path server. The access path server APS determines access timing to the end server ES which permits the access to each user terminal MN making the reported service time as a parameter. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a request reception system that efficiently suppresses a request to an end server that provides a service below a desired server system capacity, and in particular, controls the access timing to the end server according to an access path request from a user terminal. The present invention relates to a request reception system including an access path server that distributes a server load by determining each user terminal.

  When a specific event such as a television / radio program occurs, access such as a service request from a user terminal to some servers may occur in a short time and in a concentrated manner. Access concentration triggered by a specific event is likely to occur, for example, immediately after the announcement of the CD ranking in a music program or when applying for a viewer present in the program. When viewers access a specific server all at once using a communication line such as a telephone line or the Internet, the amount of access increases intensively in a short time, which may cause a response drop or system down due to server overload. .

A technique for suppressing such short time and intensive access is disclosed in Patent Document 1. In Patent Literature 1, an access path server that has received an access path request from a user terminal determines an access timing (standby time) to the end server for each received access path request, and this is a response message to each access path request. To each user terminal. The user terminal waits for the notified access timing, accesses the end server, and requests a service. The access timing is determined so that access to end servers is distributed and the access timing is in the order of accepting access path requests.
Japanese Patent Application No. 2007-207986

  In the above-described prior art, the waiting time is set and notified to each user terminal so that the number of accesses per unit time flowing into the end server does not exceed a predetermined upper limit value. The number of accesses going is controlled.

  However, the access path server cannot grasp the service provision status (for example, the number of connection sessions and service time) in the end server. For this reason, during long-term operation, there may be a difference between the processing status of the end server virtually held in the access path server and the processing status of the actual end server.

  When such a divergence occurs, it is expected that the service request more than the expected number is transmitted from the user terminal to the end server, and the end server is congested. In order to suppress the divergence, it is conceivable to collect information on the service provision status such as the number of connected sessions and service time from the end server and feed back to the access path server.

  In order to realize such feedback, it is necessary to make special modifications to the end server so that the service provision status in the end server can be observed sequentially. However, it is practically impossible to make the above modification to each of a huge number of end servers in actual operation. In addition, many end servers are owned by content providers, and the provider of the congestion control method and the end server owner are often different. That is, it is not easy to modify an end server that is the property of another person.

  The object of the present invention is to solve the above-mentioned problems of the prior art, and without having to modify the end server, accurately grasp the operation status and feed back to the access path server to optimize access to the end server. It is to provide a request reception system.

  In order to achieve the above object, the present invention notifies each user terminal of the access timing to the end server, and the user terminal transmits a service request to the end server at the notified access timing. It is characterized by the following measures.

  (1) Access timing determination means for determining the access timing to be notified to each user terminal and the service request of the sample terminal selected from the user terminal is received and transferred to the end server, and the service provided from the end server is received And a correction server for transferring to the user terminal, wherein the correction server measures the service time from the start to the end of the service provided by the end server in response to the service request, and determines the service time as an access timing. The access timing determining means determines the access timing to the end server permitted to each user terminal using the notified service time as a parameter.

  (2) The correction server further comprises means for detecting the proportion of sample terminals that have given up the service request while determining the access timing to the end server as a give-up rate, and means for notifying the give-up rate to the access timing determining means. The access timing determining means determines the access timing to the end server by reflecting the notified give-up rate.

According to the present invention, the following effects are achieved.
(1) A part of the user terminal is used as a sample terminal, and the actual service time is measured so that communication between this sample terminal and the end server relays the correction server. The service provision status can be accurately grasped and fed back to the access timing determination means.
(2) The number of sample terminals that gave up access to the end server (service request) was estimated, and this estimation result was reflected in the access timing determination in the access timing determination means, so the user who gave up the service request The capacity reserved for the terminal can be allocated to other user terminals, and as a result, the time from the access path request to the access timing can be shortened.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the best embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a main part of a network to which a request receiving system of the present invention is applied. Here, a content distribution service is requested from a user terminal MN to an end server ES (service providing server). A case will be described as an example.

  A user terminal MN such as a mobile phone, PDA or computer is connected to an IP network NW such as a mobile phone network or the Internet via a base station AP. In addition, in response to a service request from the user terminal MN, a plurality of end servers ESj (j is an end server identifier) that distributes content such as music and video, together with an access path server APS and a correction server HS described in detail later. It is connected to the IP network NW.

  The access path server APS has a function of determining a timing at which each user terminal MN is permitted to access the end server ESj and notifying each user terminal MN, and the identification information of the end server ESj that is the access destination from the user terminal MN When the access path request message including the content identification information is received, the access timing is determined based on the capability and status of the end server ESj, and this is notified to the user terminal MN. The user terminal MN waits for the notified access timing, transmits a service request to the end server ESj, and enjoys a content distribution service from the end server ESj.

  The correction server HS operates as a reverse proxy server or NAT (Network address Translation) box, relays a service request transmitted from some user terminals MN (S) operating as sample terminals to the end server ESj, The service response (content) transmitted from the end server ESj to the sample terminal MN (S) is relayed, the processing status of each end server ESj is estimated based on the reception timing of the relayed service request and service response, and the estimation result Is notified to the access path server APS.

  FIG. 2 is a functional block diagram showing the configuration of the main part of the access path server APS. In this embodiment, since the response processing executed by the access path server APS every time an access path request is received is extremely low in load, even if access path requests are received intensively in a short time, the access path server APS Access path requests can be accepted in the order received.

  The simultaneous connection session number storage unit 17 stores the maximum number of sessions Smax_j to which each end server ESj can connect simultaneously. The service time storage unit 18 stores the actual value of the session hold time necessary for providing the service desired by the user in each end server ESj as the actual service time Cj.

  The virtual queue setting unit 19 is provided to simulate the processing status of service requests in each end server ESj on the access path server APS. That is, if the maximum number of simultaneous connection sessions Smax_j of each end server ESj is n, the service request processing status in the end server ESj is n in the input stage of the end server ESj as shown in FIG. The FIFO buffer (FIFO-1, FIFO-2,..., FIFO-n) is equipped, and a service request transmitted from the user terminal NM can be simulated by being fetched from one of the FIFO buffers and sequentially processed.

  Then, register the number of queues corresponding to the number of service request accesses in the FIFO buffer, and if the transfer rate of the FIFO buffer is the reciprocal 1 / Cj of the actual service time Ci of the end server ESj, each FIFO buffer will overflow. By allocating service requests so that underflow does not occur, congestion and a decrease in efficiency of the end server ESj can be prevented.

  Therefore, in the present invention, a virtual queue VQ simulating each FIFO buffer described above is provided in the access path server APS, and service requests are accessed for each access path request so that neither overflow nor underflow occurs in the virtual queue VQ. The above technical problem is solved by assigning timing.

  Therefore, the virtual queue setting unit 19 is set with the same number of virtual queues VQk, j (k is a virtual queue identifier) as the maximum number of simultaneous connection sessions Smax_j for each end server ESj. In each virtual queue VQk, j, as will be described in detail later, a queue is registered (enqueued) every time an access path request transmitted from the user terminal MN is received, and the registered queue is stored in the actual service time Ci. Discarded (dequeued) sequentially over time at a rate corresponding to the inverse 1 / Cj. Thereby, in the present embodiment, the processing status of the service request in the end server ESj can be simulated as the queue length of each virtual queue VQk, j on the access path server APS.

  The access path request receiving unit 10 receives an access path request transmitted from each user terminal MN via an interface (I / F). The request analysis unit 11 analyzes the received access path request, and identifies the end server ESj that is the service request destination and the requested service. The virtual queue selection unit 12 selects the virtual queue VQkx, j having the shortest queue length from the plurality of virtual queues VQk, j associated with the access destination end server ESj specified by the access path request. Thereafter, the subscript x is attached to the virtual queue having the smallest queue length and the identifier k of the queue value.

  The virtual queue updating unit 13 adds the queue registered in the virtual queue VQk, j of each end server ESj together with the adding unit that enqueues to the selected virtual queue VQkx, j based on the actual service time Cj and the elapsed time. And a discarding unit for sequentially dequeuing.

  As will be described in detail later, the access timing determination unit 14 determines an access timing for responding to the access path request based on the queue length VQkx, j [t] of the selected virtual queue VQkx, j. In this embodiment, the access timing is determined as a standby time from the current time t. The access path response generation unit 15 generates an access path response including the access timing to the end server ESj. The access path response reply unit 16 replies the access path response to the user terminal that transmitted the access path request via the I / F.

  FIG. 4 is a functional block diagram showing the configuration of the main part of the correction server HS. The receiving unit 31 includes a service request receiving unit 31a and a service response receiving unit 31b, and receives a service request transmitted from the user terminal MN and a service response transmitted from the end server ESj via the I / F. The transmission unit 32 includes a service request transmission unit 32a and a service response transmission unit 32b, and transmits a service request to be transferred to the end server ESj and a service response to be transferred to the user terminal MN via the I / F.

  The service start detection unit 33 detects the service start timing based on the received service request packet. The service end detection unit 34 detects the service end timing based on the received service response packet. The timer 35 measures the current time t0. The service time measuring unit 36 measures the actual service time Cj in the end server ESj based on the elapsed time from the service start timing to the service end timing for the same session.

  The give-up rate calculation unit 37 is based on the ratio of sample terminals MN (s) that do not transmit an access request at this access timing even though the access timing of the service request is assigned from the access path server APS. The proportion of user terminals MN that gave up the service request, that is, the “giving up rate” is estimated.

  The monitoring result notifying unit 38 uses the actual service time Cj of each end server ESj measured by the service time measuring unit 36 and the give rate Rj for each end server ESj calculated by the give rate calculating unit 37 to the access path server APS. To notify.

  Returning to FIG. 2, the monitoring result receiving unit 18 of the access path server APS receives the actual service time Cj and the give-up rate Rj of each end server ESj notified from the correction server HS. The service time update unit 20 updates the actual service time Cj stored in the service time storage unit 18 to the notified actual service time Cj. The notified give-up rate Rj is notified to the correction unit of the virtual queue update unit 13. The correction unit corrects the queue value of the virtual queue VQj based on the notified give-up rate Rj. A method for correcting the cue value based on the give-up rate Rj will be described in detail later.

  Next, the operation of the present invention will be described in detail with reference to a flowchart. FIG. 5 is a flowchart showing a procedure of an access path request and a service request in the user terminal MN requesting content distribution, FIG. 6 is a flowchart showing a procedure of an access path response in the access path server APS, and FIG. It is a sequence flow. In addition, the definition of each code | symbol in this embodiment is as follows.

i: Access path request identifier
j: End server ES identifier
k: Virtual queue VQ identifier
sti: Eigenvalue of service time required to provide the service requested by access path request i (unique service time)
ti, j: i-th access path request time for end server ESj
di, j: Delay time until the access timing assigned to the i-th access path request for the end server ESj
Cj: Actual service time for each end server ESj (actual service time)
Qmax_j: Maximum queue value for each end server ESj
bi, j: Weight value of the requested service
VQj (ti, j)-: Queue length of the end server ESj immediately before receiving the access path request at time ti
VQj (ti, j) +: End server ESj queue length immediately after updating to reflect the access path request received at time ti

  The user operates a key switch or the like of his / her user terminal MN to execute a content request operation, and when this is detected in step S11 of FIG. 5, the process proceeds to step S12. In step S12, an access path request destined for the access path server APS including the identifier of the requested content and the identifier j of the end server ES that provides the content is generated and transmitted in step S13.

  In the access path server APS, when the access path request is received by the access path request receiving unit 10 in step S31 of FIG. 6, the process proceeds to step S32. In step S32, the request analysis unit 11 analyzes the access path request, and identifies the requested content and the requested end server ESj. In subsequent steps S33 to S35, the queue values at the current time t of the plurality of virtual queues VQk, j assigned to the requested end server ESj are the elapsed time Dt from the previous update time ti-1 and the end. It is updated based on the actual service time Cj in the server ESj.

  That is, in step S33, each virtual queue VQk, immediately after updating the access path request received at time ti-1 for the end server ESj corresponding to the identifier j registered in the received access path request. The queue length [VQk, j (ti-1) +] of j is taken from the virtual queue setting unit 19. In step S34, the actual service time Cj in the end server ESj is fetched from the service time storage unit 18. In step S35, the queue length [VQk, j (ti) −] of each virtual queue VQk, j at the reception time ti of the current access path request is obtained by the following equation (2).

  VQk, j (ti)-= VQk, j (ti-1) +-(ti-ti-1) / Cj (2)

  In step S36, the virtual queue VQkx, j having the smallest queue length [VQk, j (ti) −] is selected as a queue registration destination for the current access path request. In step S37, the access timing notified to the user terminal MN that transmitted the access path request is obtained by the following equation (3) as a delay time di, j with reference to the reception time ti of the access path request.

  di, j = VQkx, j (ti) − × Cj (3)

  In step S38, the queue value [VQkx, j (ti) +] of the selected virtual queue VQkx, j is calculated based on the following equation (4), and the queue value [VQk, j of the other virtual queue VQk, j is calculated. j (ti) +] is calculated based on the following equation (5). In step S39, the queue value [VQk, j (ti) +] of each virtual queue VQk, j is updated based on the calculation result.

  VQkx, j (ti) + = VQkx, j (ti) − + bi (4)

  VQk, j (ti) + = VQk, j (ti) − (5)

  Here, the weight value [bi] is a queue value representing the specific service time sti necessary for providing the requested service, and bi = 1 if the specific service time sti is equal to the actual service time Cj. If the unique service time sti is longer than Cj, bi> 1; if the unique service time sti is shorter than Cj, bi <1.

  In step S40, the user terminal MN that transmitted the current access path request is classified as a sample terminal MN (S) with a predetermined probability Rs. In this embodiment, a random number rnd of 0 <rnd ≦ 1 is generated. If rnd ≦ Rs, the random number rnd is classified as a sample terminal MN (S) and the process proceeds to step S41, and one of the consecutive sample IDs is assigned in ascending order.

  In step S42, the access path response generator 15 generates an access path response including the delay time di, j as the access timing. At this time, the sample ID is further registered in the access path response addressed to the sample terminal MN (S). In step S43, the access path response is returned from the access path response reply unit 16 to the transmission terminal of the access path request.

  Returning to FIG. 5, when the user terminal MN receives the access path response in step S14, the delay time di, j registered in the access path response is extracted in step S15. In step S16, the extracted delay time di, j is compared with a predetermined upper limit value dmax. If the delay time di, j> dmax, the process proceeds to step S17, and an error message indicating that the service cannot be provided is displayed on the terminal display. Then, the processing is stopped. On the other hand, if the delay time di, j ≦ dmax, the process proceeds to step S18, and as shown in FIG. 8 (a) and FIG. 8 (b), the reception end indicating that the requests are processed in the order of arrival. A message appears on the terminal display.

  In step S19, a service request message including the identifier of the content requesting distribution and the identifier of the end server and having the request destination end server ESj as the destination address is generated. In step S20, it is determined whether or not it is a sample terminal based on whether or not the sample ID is notified from the access path server APS. If it is a sample terminal, the process proceeds to step S21. In step S21, the destination address of the service request is rewritten to the address of the correction server HS, and the notified sample ID is added to a predetermined field of the service request. In the example shown in FIG. 7, the user terminal MN3 is a sample terminal, and the user terminals MN1 and MN2 are non-sample terminals.

  In step S22, it is determined whether or not it is the access timing based on whether or not the elapsed time since the transmission of the access path request in step S13 has reached the delay time di, j, and waits for the access timing. Then, the process proceeds to step S23. In step S23, the service request generated in step S19 is transmitted. Such standby processing up to the access timing in the user terminal MN can be implemented by using various script languages such as Java (registered trademark) script and Flash.

  The end server ESj that has received this service request prepares the requested content and distributes it to the user terminal MN. When receiving the content in step S24, the user terminal MN proceeds to step S25 and stores the content. In step S26, a download completion message shown as an example in FIG. 9 is displayed on the terminal display.

  FIG. 10 is a flowchart showing the procedure of the giving rate (Rj) calculation process executed by the correction server HS, and mainly shows the operation of the giving rate calculating unit 37 in FIG. The give-up rate (Rj) calculation process is repeatedly executed continuously in a predetermined cycle.

  In step S51, a timer Tr that times out in a predetermined unit time ΔT starts, and when the service request transmitted from the sample terminal MN (S) is received in step S52, the process proceeds to step S53. In step S53, the sample ID registered in the received service request is extracted and temporarily stored in step S54. In step S55, it is determined whether or not the timer Tr has timed out, and the process returns to step S52 until the time out. Since the processing in steps S52 to S55 is performed for each end server Ej that is the final destination of the service request, the extracted sample ID is stored for each end server Ej that is the final destination.

  Thereafter, when a time-out of the timer Tr is detected in step S55, the process proceeds to step S56, and based on the sample ID extracted in the unit time ΔT, the sample terminal assigned the unit time ΔT as the access timing of the service request The total number Nsmp of MN (S) is estimated for each end server Ej. In this embodiment, the difference (IDmax1-IDmax0) between the maximum value IDmax0 of the sample ID extracted in the previous unit time ΔT and the maximum value IDmax1 of the sample ID extracted in the current unit time ΔT is the total number of samples Nsmp. It is said.

  In step S57, the number of sample terminals MN (S) giving up the service request is obtained by subtracting the total number Nrcv of sample IDs actually extracted within the unit time ΔT from the total number of samples Nsmp. The give-off rate Rj is calculated by dividing by Nsmp. In step S58, the give-up rate Rj is notified to the access path server APS.

  In the present embodiment, the access path server APS has been described as selecting the sample terminal MN (S). However, the present invention is not limited to this, and the user terminal MN attempting to transmit an access path request. May set itself as the sample terminal MN (S) with a predetermined probability. At this time, if the sample terminal MN (S) notifies the correction server HS of its own identifier as a sample ID in advance, the correction server HS could not receive the number of terminals that notified the sample ID in advance and the service request. The give-up rate Rj can be calculated in the same manner based on the ratio with the number of terminals.

  FIG. 11 is a flowchart showing the procedure of the virtual queue correction process executed by the access path server APS in response to the notification of the give-up rate Rj, and mainly the operation of the correction unit in the virtual queue update unit 13 of FIG. Is shown.

  In step S71, when the notification of the give-off rate Rj related to the end server ESj is detected, the process proceeds to step S72, where the total queue value ΣVQj (t) of all the virtual queues VQj (t) at the current time t of the end server ESj is calculated. Desired. In step S73, the product of the total queue value ΣVQj (t) and the give-up rate Rj is calculated as the number of queues to be discarded (dequeue number). In step S74, queues corresponding to the number of dequeues are equally dequeued (discarded) from each virtual queue VQ of the end server ESj.

  FIG. 12 is a diagram schematically showing the contents of the virtual queue correction process. Here, five virtual queues VQ1, j to VQ5, j are set in the end server ESj, and each virtual queue VQ1, j is shown in FIG. There are a total of 20 queues registered in ~ VQ5, j. Therefore, if the notified give-up rate Rj is 20%, the number of dequeues is “4”, so that the virtual queues are discarded one by one from the four virtual queues VQ1, j to VQ4, j.

  13, 14, and 15 are diagrams schematically showing a procedure of actual service time calculation processing executed each time the correction server HS receives and relays a service request from the sample terminal MN (S). When the service response is performed using a TCP session, the actual service time Cj is obtained by monitoring the SYN / FIN packet, HTTP METHOD, HTTP message body, or HTTP / 1.1 range request.

  FIG. 13 is a diagram schematically illustrating a method for estimating the service start time and the service end time based on the monitoring result of the SYN / FIN packet.

  The TCP session is composed of a plurality (N) of TCP connections, and each TCP connection returns a SYN-ACK to the SYN packet transmitted from the sample terminal MN (S), and the sample terminal MN ( It is established by transmitting an ACK packet from S) to the end server ESj. Then, the TCP connection established in this way is disconnected when both the sample terminal MN (S) and the end server ESj exchange FIN-ACK.

  Therefore, in this embodiment, the time at which the first SYN-ACK transmitted from the end server ESj is received by the correction server HS is set as the service start time, and further, from the sample terminal MN (S) in the Nth FIN-ACK exchange. The time at which the sent FIN-ACK is received by the correction server HS is the service end time, and the service time of each service request is from the service start time to the service end time. The service time is Cj.

  Since the number of TCP connections (N) depends on the service content, it is desirable that the correction server HS obtains the number of TCP connections (N) for each service from each end server ES in advance.

  Whether each TCP connection is related to the same TCP session can be determined based on, for example, whether the source IP address and port number are the same, and the destination IP address and port number are the same. . Alternatively, the determination may be made based on whether the IDs unique to the transmission source terminal are the same and the destination IP address and the port number are the same.

  FIG. 14 is a diagram schematically representing a method of estimating the actual service time Cj based on the monitoring result of HTTP METHOD.

  In the present embodiment, a TCP connection is established between the sample terminal MN (S) and the end server ESj when an HTTP METHOD (for example, GET, POST, etc.) transmitted from the sample terminal MN (S) passes through the correction server HS. Is determined as the service start time.

  Furthermore, the service end time is the time when the FIN-ACK transmitted from the sample terminal MN (S) in the Nth FIN-ACK exchange is received by the correction server HS, and each service request is from the service start time to the service end time. The arithmetic average value or the moving average value is the actual service time Cj.

  FIG. 15 is a diagram schematically representing a method for estimating the actual service time Cj based on the monitoring result of the HTTP message body.

  In the present embodiment, the HTTP message body transmitted from the sample terminal MN (S) is monitored by the correction server HS, and an identification code (for example, a character string such as “start”) is set in a predetermined area of the message body. ) Is registered, the time is set as the service start time.

  Furthermore, when it is detected that an identification code (for example, a character string such as “finish”) is registered in a predetermined area of the HTTP message body transmitted from the sample terminal MN (S), This time is the service start time. The service time of each service request is from the service start time to the service end time, and the arithmetic average value or moving average value thereof is the actual service time Cj.

  In HTTP / 1.1, it is possible to download only the designated part of the content by a range request when acquiring the content. For example, when downloading a certain content, if “Range: bytes = x-y” is described in the HTTP header, the x-th to y-th bytes of the content can be selectively acquired.

  On the other hand, a header including “Content-Length” and “Content-Range” is returned from the end server ESj that has received such a range request. "Content-Length" describes the requested data size from the x-th byte to the y-th byte, and "Content-Range" describes the requested data range and the data size of the entire content ing.

  In the correction server HS, the time when the packet described with “Range: bytes = 0-y” in the HTTP header is received from the sample terminal MN (S) is the service start time. The total data size registered in “Content-Range” is stored as the remaining download size, and every time a data packet is received from the end server ESj, the data described as “Content-Length” in its header The time when the size is subtracted from the residual size and the residual size becomes zero is taken as the service time, and the arithmetic average value or moving average value thereof is taken as the actual service time Cj.

  The actual service time Cj of each end server ESj obtained as described above is notified to the access path server APS. In the access path server APS, the service time storage unit 18 that stores the actual service time Cj of each end server is updated based on the notified actual service time Cj.

  In the above embodiment, the access path server APS receives the access path request from the user terminal MN, determines the access timing to the end server ES that is allowed to each user terminal MN, and uses this as the access path for each user terminal. Although described as being notified to the MN, the present invention is not limited to this, and the access timing may be notified to each user terminal MN by a broadcast wave.

  FIG. 16 is a block diagram showing a configuration of a main part of the network in which access timing is notified to each user terminal MN by broadcast waves.

  The control server CS has an access timing determination function for determining an access timing to be notified to each user terminal MN. Similarly to the access path server APS, the control server CS determines the access timing to each end server ESj based on the actual service time Cj and the give-off rate Rj of each end server ESj notified from the correction server HS. To the broadcasting station 1. The broadcast station 1 broadcasts the notified access timing to each end server ESj superimposed on the broadcast wave. Therefore, the access timing to each end server ESj superimposed on the broadcast wave is updated every time the access timing is newly notified from the control server CS to the broadcast station 1.

  In each user terminal MN, a broadcast wave is received in response to a predetermined service request operation by the user, and by extracting the access timing superimposed on this, the access timing to each end server ESj at that time is determined. recognize. Thereafter, each user terminal MN transmits a service request after waiting for the access timing.

  At this time, each user terminal MN performs a process of setting itself to the sample terminal MN (S) with a predetermined probability for each service request destination end server ESj, and the sample terminal MN (S) is the same as described above. The end server ESj is accessed via the correction server HS. The correction server HS calculates the actual service time Cj and the give-up rate Rj for each sample terminal MN (S) and end server ESj, and notifies the control server CS of this.

It is a block diagram of a network to which the request reception system of the present invention is applied. It is a functional block diagram showing a configuration of a main part of an access path server (APS). It is the figure which showed the function of the virtual queue setting part typically. It is a functional block diagram showing a configuration of a main part of an access path server (HS). It is the flowchart which showed the request | requirement procedure by a user terminal. It is the flowchart which showed the procedure of the response in an access path server. It is a sequence flow among a user terminal, an access path server, a correction server, and an end server. It is the figure which showed an example of the reception completion message. It is the figure which showed an example of the download completion message. It is the flowchart which showed the procedure of the give-up rate calculation process. It is the flowchart which showed the procedure of the virtual cue correction processing. It is the figure which expressed virtual cue correction processing typically. It is the figure which showed typically the measuring method (the 1) of performance service time. It is the figure which showed typically the measuring method (the 2) of performance service time. It is the figure which showed typically the measuring method (the 3) of performance service time. It is a block diagram of a network in which access timing to each end server is notified to each user terminal MN by broadcast waves.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Broadcasting station, 10 ... Access path request receiving part, 11 ... Request analysis part, 12 ... Virtual queue selection part, 13 ... Virtual queue update part, 14 ... Access timing determination part, 15 ... Access path response generation part, 16 ... Access path response reply unit, 17 ... concurrent connection session number storage unit, 18 ... service time storage unit, 19 ... virtual queue setting unit, 31 ... reception unit, 32 ... transmission unit, 33 ... service start detection unit, 34 ... service end Detection unit, 36 ... service time measurement unit, 37 ... give-up rate calculation unit, 20 ... service time update unit

Claims (8)

In the request reception system that notifies the access timing to the end server to each user terminal, the user terminal transmits a service request to the end server at the notified access timing,
Access timing determining means for determining an access timing to be notified to each user terminal;
Receiving a service request of the sample terminal selected from the user terminal and transferring it to the end server, and a correction server receiving the service provided from the end server and transferring it to the user terminal,
The correction server is
Means for measuring the service time from the start to the end of the service provided by the end server in response to the service request;
Means for notifying the service time to the access timing determination means,
The access timing determination unit determines an access timing to an end server permitted to each user terminal using the notified service time as a parameter. Means for transmitting the access timing determined by the access timing determination means by a broadcast wave;
The request reception system according to claim 1, wherein the user terminal includes means for extracting an access timing from the broadcast wave in response to a predetermined operation.   The access timing determination means is an access path server that receives an access path request from a user terminal, determines an access timing to an end server that is permitted to each user terminal, and notifies this to each user terminal as an access path. The request reception system according to claim 1. The correction server further includes:
Means for detecting the proportion of sample terminals that gave up the service request while determining the access timing to the end server as a give-up rate;
Means for notifying the give-up rate to the access timing determination means,
The request reception system according to claim 1, wherein the access timing determination unit determines an access timing to the end server by reflecting the notified give-up rate. The access path server is
Means for storing the actual service time value in the end server as the actual service time;
Means for setting a plurality of virtual queues in which queues representing the number of accesses to the end server are virtually registered based on the number of simultaneous connection sessions of the end server;
Means for receiving an access path request from a user terminal;
Means for discarding the queue from each virtual queue based on elapsed time and actual service time at the end server, and adding the queue to the virtual queue in response to an access path request;
Means for determining access timing to the end server of the user terminal based on a queue value of the virtual queue to which the queue is added;
Means for notifying the user terminal of the determination result of the access timing,
The request reception system according to claim 3 or 4, wherein the service time notified from the correction server is updated and registered as the actual service time. The access path server is
Means for storing the actual service time value in the end server as the actual service time;
Means for setting a plurality of virtual queues in which queues representing the number of accesses to the end server are virtually registered based on the number of simultaneous connection sessions of the end server;
Means for receiving an access path request from a user terminal;
Means for discarding queues from each virtual queue based on elapsed time and actual service time at the end server, and adding queues to the virtual queue in response to access path requests;
Means for determining access timing to the end server of the user terminal based on a queue value of the virtual queue to which the queue is added;
Means for notifying the user terminal of the determination result of the access timing,
6. The request reception system according to claim 5, wherein a queue is discarded from each of the virtual queues based on the give-up rate notified from the correction server. With multiple end servers,
The request reception system according to claim 5 or 6, wherein the plurality of virtual queues are set for each end server. The access path server selects a sample terminal with a predetermined probability from the user terminal that has received the access path request, notifies the sample terminal of the access timing and sample ID,
The user terminal other than the sample terminal transmits a service request to the end server at the notified access timing,
The request reception system according to claim 3, wherein the sample terminal transmits a service request to the correction server at the notified access timing.

Images