Congestion control system and method for web service

• US 20030099197A1
• Filed: 07/09/2002
• Published: 05/29/2003
• Est. Priority Date: 11/28/2001
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Hitachi, Ltd.
• Inventors: Takahashi, Yasuhiro, Yokota, Daisuke, Noda, Fumio
• Granted Patent: US 7,239,609 B2
• Time in Patent Office: Days
• US Class Current: 370/230
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/11 Identifying congestion
• CPC Class Codes: H04L 47/15 in relation to multipoint traffic arrangements for broadcast or multicast in data networks H04L12/18
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/28 in relation to timing considerations
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/745 Reaction in network
• CPC Class Codes: H04L 47/822 Collecting or measuring resource availability data
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

A congestion control system provided in a connection portion between a network for an Internet provider and the Internet estimates a processing capability of a Web server on the basis of states of TCP/IP such as a window size of connection being connected, a cancellation rate of packet, a response time, a throughput and the like in a congestion state of the Web server to be requested and a network to the Web server and calculates a regulation duration for each Web server on the basis of the processing capability and the number of clients being in a waiting state for a re-request to the Web server so that the client is regulated when a response for regulating the re-request with the regulation duration is returned to the client. Consequently, since the number of re-requests per unit time can be made fixed, congestion due to re-requests can be prevented and the user can necessarily utilize the Web server upon re-request. Further, the service utilizable hour can be grasped to utilize the Web service deliberately.

Claims

1. A congestion control system used in the Web service in which a user utilizes the Web to receive contents and provided between user'"'"'s Web clients and Web servers for providing contents, comprising:

means for relaying a contents receiving request from the Web client and a response of contents to said contents receiving request; and

means for transmitting as a response a connection request regulation message for regulating the Web client from transmitting the contents receiving request to the Web server for a predetermined connection regulation duration to the Web client instead of the contents when the relay state at the time of the relaying is detected to be in a previously set congestion state.

2. A congestion control system according to claim 1, comprising:

identification means for identifying a Web server storing contents corresponding to said contents receiving request on the basis of said contents receiving request;

responsive state detection means for detecting a responsive state in the TCP/IP layer from said identified Web server; and

totalization means for totalizing said detected responsive state, the number of contents receiving requests from other users to said identified Web server and the number of connection request regulation messages transmitted to the Web client as a response with regard to said identified Web server;

whereby the connection regulation duration is controlled on the basis of said totalization and the connection request regulation message is transmitted to said Web client as a response.

3. A congestion control system according to claim 1, comprising:

positional information obtainment means for obtaining positional information added by the Web client from said contents receiving request;

division means for dividing Web clients into each area on the basis of said obtained positional information;

reception state detection means for detecting a reception state in the TCP/IP layer of said Web client with regard to transmission of contents to said divided Web client; and

totalization means for totalizing the detected reception state, the number of Web clients in a specific divided area and the number of connection request regulation messages transmitted to said specific divided area as responses;

whereby the connection regulation duration is controlled on the basis of said totalization and the connection request regulation message is transmitted to said Web client as a response.

4. A congestion control system according to claim 1, wherein

said connection request regulation message is transmitted with a connection regulation duration different to said Web client as a response.

5. A method of relaying a contents receiving request transmitted from a Web client and a response of contents from a predetermined Web server corresponding to said contents receiving request, comprising:

transmitting as a response a connection request regulation message for regulating said Web client from transmitting the contents receiving request to said Web server for a predetermined connection regulation duration to the Web client instead of the contents when the relay state at the time of the relaying is detected to be in a previously set congestion state, said connection regulation duration being not a fixed value.

6. A relaying method according to claim 5, wherein

said connection regulation duration is varied in accordance with a congestion state of the Web server.

7. A relaying method according to claim 6, wherein

said connection regulation duration is varied in accordance with information concerning processing capability of the Web server and information concerning the number of current contents receiving requests.

8. A relaying method according to claim 6, wherein

said connection regulation duration is varied in accordance with information concerning the Web client.

9. A relaying method according to claim 5, wherein

said connection regulation duration is varied in accordance with a congestion state in network between the Web server and the Web client.