CN103560937A - Data accessing method and entry server - Google Patents

Abstract

The present application discloses a data access method and an entry server, wherein the method includes: the entry server receives access request information to be transmitted, and the access request information is used to access data on the destination server; The link information among them is dynamically assigned by the configuration system; the entry server sends the access request information to the destination server according to the link information. The invention solves the technical problem that the forwarding port of the existing TCP proxy server is fixed in advance and can only be manually modified by the user, but cannot be dynamically modified. The dynamic allocation of the link increases the utilization rate of the link and speeds up the process. the speed of data transmission.

Description

Data Access Methods and Ingress Servers

technical field

This application relates to the Internet field, and in particular, relates to a data access method and an entry server.

Background technique

In the CDN field, by directing the user's access to the nearest lower-layer TCP proxy server, the lower-layer TCP proxy server accepts the access request from the user by listening to a specific socket (IP address + port), and uses the pre-configured The socket of the upper layer TCP proxy server establishes a connection with the upper layer, then forwards the request to the upper layer, and forwards the response from the upper layer to the user, without analyzing locally.

By building multiple TCP proxy servers on the Internet, through the forwarding of multi-layer TCP proxy servers, users can access the source site, and use high-quality links between TCP proxy servers to realize user accelerated access.

The existing common technology is the TCP proxy forwarding mode. The listening socket of the lower-layer TCP proxy server forwards the TCP request to the upper-layer proxy server, and the upper-layer TCP proxy server forwards the TCP request to the origin server. But these sockets must be pre-configured in the TCP proxy server in advance, and cannot be automatically modified according to the link situation, and can only be manually modified by the user.

For the above problems, no effective solution has been proposed yet.

Contents of the invention

The main purpose of this application is to provide a data access method and an entry server to at least solve the problem that the forwarding port of the existing TCP proxy server is fixed in advance and can only be manually modified by the user, and cannot be dynamically modified.

According to one aspect of the present application, a data access method is provided, which includes: the entry server receives access request information to be transmitted, wherein the access request information is used to access data on the destination server; The link information between them, where the link information is calculated by the configuration system according to the delay information and packet loss rate information from itself to other network servers in the network server set reported by the network server including the entry server and the destination server ; The entry server sends the access request information to the destination server according to the link information.

As an optional solution, before the ingress server receives the access request information to be transmitted, the data access method further includes: each network server in the network server set including the ingress server and the destination server every predetermined period or when meeting Obtain the delay information and packet loss rate information from itself to other network servers in the network server set under predetermined conditions; the network server sends the delay information and packet loss rate information to the configuration system; the entry server receives the entry server returned by the configuration system The link information with the destination server: the entry server obtaining the link information between the entry server and the destination server includes: the entry server locally obtaining the link information between the entry server and the destination server.

As an optional solution, the predetermined condition includes: the network server receives a link information acquisition request sent by the configuration system.

As an optional solution, the ingress server acquiring link information between the ingress server and the destination server includes: configuring the system to receive a first request message for acquiring link information sent by the ingress server; configuring the system to respond to the first request message Send a second request message for obtaining link information to each network server in the network server set including the entry server and the destination server; the configuration system receives the information sent by the network server to other network servers in the network server set. Delay information and packet loss rate information; the configuration system calculates the link information between the ingress server and the destination server according to the delay information and packet loss rate information; the configuration system sends the link information between the ingress server and the destination server to the ingress server.

As an optional solution, the ingress server sending the access request information to the destination server according to the link information includes: the ingress server sends the access request information to the egress server connected to the destination server through multiple intermediate servers indicated by the link information; The server sends the access request information to the destination server; wherein, transmitting the access request information between multiple intermediate servers includes the following steps: the first intermediate server in the multiple intermediate servers obtains the link information as the next link of the first intermediate server. Jump the socket of the second intermediate server; the first intermediate server establishes a connection with the second intermediate server according to the socket of the second intermediate server; if the connection between the first intermediate server and the second intermediate server is successfully established, the first intermediate server The server deletes the socket of the first intermediate server from the link information, and sends the updated link information and access request information to the second intermediate server; wherein, the socket is an IP address and a port number or is an IP address .

As an optional solution, the transmission of the access request information between multiple intermediate servers includes: if the connection between the first intermediate server and the second intermediate server fails to be established, the first intermediate server obtains from the link information as the second intermediate server The socket of the third intermediate server of the next hop; the first intermediate server establishes a connection with the third intermediate server according to the socket of the third intermediate server; if the connection between the first intermediate server and the third intermediate server is successfully established, then the first intermediate server An intermediate server deletes the sockets of the first intermediate server and the second intermediate server from the link information, and sends the updated link information and access request information to the third intermediate server.

According to another aspect of the present application, an entry server is provided, which includes: a first transmission unit, configured to receive access request information to be transmitted, wherein the access request information is used to access data on the destination server; an acquisition unit, Used to obtain the link information between the entry server and the destination server, where the link information is provided by the configuration system according to the delay from itself to other network servers in the network server set reported by the network servers including the entry server and the destination server information and packet loss rate information are calculated; the second transmission unit is configured to send the access request information to the destination server according to the link information.

As an optional solution, the portal server includes: a third transmission unit, configured to acquire the portal server to other information in the network server set every predetermined period or when a predetermined condition is met before receiving the access request information to be transmitted. Delay information and packet loss rate information of the network server, wherein the set of network servers includes an entry server and a destination server; the delay information and packet loss rate information are sent to the configuration system; the fourth transmission unit is used to receive the information returned by the configuration system The link information between the entry server and the destination server; wherein the obtaining unit is used to locally obtain the link information between the entry server and the destination server.

As an optional solution, the predetermined condition includes: the network server receives a link information acquisition request sent by the configuration system.

As an optional solution, the acquisition unit includes: a first transmission module, configured to configure the system to send a first request message for acquiring link information; a second transmission module, configured to receive the configuration system to send in response to the first request message The second request message used to obtain link information; the third transmission module, used to send the delay information and packet loss rate from the ingress server to other network servers in the network server set to the configuration system in response to the second request message information, wherein the set of network servers includes an ingress server and a destination server; the fourth transmission module is used to receive the link information between the ingress server and the destination server calculated according to the delay information and packet loss rate information sent by the configuration system.

Through the technical solution of the present application, the following beneficial effects can be achieved:

1) Perceive the information of the entire link globally, and calculate the link based on the link information obtained from each network server, so that link resources can be used more reasonably, the bandwidth utilization of the entire network can be maximized, and the number of links can be increased. utilization rate;

2) The socket used to establish a connection is parsed from the dynamically allocated link information without pre-configuration, thereby reducing the complexity of configuration and speeding up data transmission;

3) The IP of the entry server can be brought to the destination server, so that the destination server can obtain the IP of the netizen.

Of course, implementing any product of the present application does not necessarily need to achieve all the above-mentioned advantages at the same time.

Description of drawings

The drawings described here are used to provide a further understanding of the application and constitute a part of the application. The schematic embodiments and descriptions of the application are used to explain the application and do not constitute an improper limitation to the application. In the attached picture:

FIG. 1 is an optional schematic diagram of a transmission network according to an embodiment of the present application;

FIG. 2 is an optional flow chart of a data access method according to an embodiment of the present application;

FIG. 3 is another optional schematic diagram of a transmission network according to an embodiment of the present application;

FIG. 4 is another optional flow chart of a data access method according to an embodiment of the present application;

FIG. 5 is another optional flowchart of a data access method according to an embodiment of the present application;

FIG. 6 is another optional flowchart of a data access method according to an embodiment of the present application;

FIG. 7 is another optional schematic diagram of a transmission network according to an embodiment of the present application;

FIG. 8 is another optional schematic diagram of a transmission network according to an embodiment of the present application;

FIG. 9 is an optional schematic diagram of an entry server according to an embodiment of the present application;

FIG. 10 is another optional schematic diagram of an ingress server according to an embodiment of the present application; and

Fig. 11 is another optional schematic diagram of an ingress server according to an embodiment of the present application.

Detailed ways

Hereinafter, the present application will be described in detail with reference to the drawings and embodiments. It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

As used in this application, the term "module", "component" or "unit" may refer to a software object or routine executed on a computer. Although the systems and methods described herein are implemented in software, implementations in hardware or a combination of software and hardware are also possible and contemplated.

Example 1

As shown in FIG. 1 , the transmission network shown in this embodiment includes: entry servers 102 - 1 and 102 - 2 , intermediate servers 106 - 1 to 106 - 3 , and destination server 104 . The entry servers 102-1 and 102-2 may send the data access request to the corresponding destination server 104 through one or more of the intermediate servers 106-1 to 106-3. It should be noted that the transmission network shown in FIG. 1 is only an example, and the network topology among the ingress server, the intermediate server, and the destination server can also be other connection modes, and the number of intermediate servers is not limited in this embodiment. Limitation, as an optional implementation, the entry server can also be directly connected to the destination server.

Based on the transmission network shown in Figure 1, this embodiment also provides an optional data access method, which is described below using the entry server 102-1 as an example, as shown in Figure 2, the data access method includes but is not limited to Follow the steps below:

S202, the entry server 102-1 receives access request information to be transmitted, wherein the above access request information is used to access data on the destination server 104;

Optionally, the above access request information may include a web page address input by the user on the terminal, for example, www.baidu.com. In addition, different portal servers can be set for different regions. For example, when a user in Guangzhou enters a webpage address on the terminal, the access request information carrying the webpage address can be routed to the portal server 102-1 set in Guangzhou. superior.

S204, the entry server 102-1 obtains the link information between the entry server 102-1 and the destination server 104, wherein the link information is reported by the configuration system to the network server according to the network server including the entry server and the destination server Delay information and packet loss rate information of other network servers in the set are calculated;

As an optional manner, the link information between the ingress server 102-1 and the destination server 104 includes optimal link information, and optionally, the optimal link information may be a link with the least number of hops.

As an optional method, obtaining link information may include the following steps: obtaining link information to the entry server 102-1, intermediate servers 106-1 to 106-3, and destination server 104, and calculating Link information between the entry server 102-1 and the destination server 104. Here, the link information of the ingress server 102-1, the intermediate servers 106-1 to 106-3, and the destination server 104 may include but not limited to: delay information and packet loss rate information between each server and other network servers.

Optionally, it is assumed that the link between the entry server 102-1 and the destination server 104 is: entry server 102-1—intermediate server 106-1—intermediate server 106-2—destination server 104. For the above-mentioned link, the above-mentioned link information may include but not limited to: the socket of the intermediate server 106-1, the intermediate server 106-2, and the intermediate server 106-2. Here, the socket includes: IP address and port number or IP address.

S206, the entry server 102-1 sends the above access request information to the above destination server 104 according to the above link information.

Optionally, the entry server 102-1 may transmit to the destination server 104 through the intermediate server 106-1 and the intermediate server 106-2, or transmit to the destination server 104 through the intermediate server 106-1 and the intermediate server 106-3. It should be noted that the above transmission path is only an example, and for different network topologies, different transmission paths may exist between the ingress server 102 - 1 and the destination server 104 .

As an optional manner, after the above access request is completed, the link connection established by the entry server 102-1 to the destination server 104 will also be removed.

As another alternative, after the ingress server 102-1 obtains the link information between the ingress server 102-1 and the destination server 104, the ingress server 102-1 will pre-establish multiple complete link connections, and put In the connection pool, when an access request is accessed, the entry server 102-1 can take an established connection from the connection pool to transmit data, so that the time for establishing a connection can be reduced and the effect of accelerating transmission can be achieved; After the above access request is completed, the link connection established by the entry server 102-1 to the destination server 104 will also be removed. A new connection will be added to the connection pool immediately for subsequent access requests. In addition, all connections have a timeout period, which is determined by the destination server 104. Connections within the connection timeout period are available, and when the connection timesout, the connection will be actively removed from the connection pool, and the connection pool is always guaranteed Connection is available.

Through the above-mentioned embodiment, the information of the entire link is perceived globally, and the link is calculated according to the acquired link information perceived by each network server, so that link resources can be used more reasonably, and the bandwidth utilization of the entire network can be maximized. Increase link utilization; in addition, the socket used to establish a connection is parsed from dynamically allocated link information without pre-configuration, thereby reducing the complexity of configuration and speeding up data transmission ; Further, the IP of the entry server can be brought to the destination server, so that the destination server can obtain the IP of the netizen.

In order to obtain dynamically allocated link information, this embodiment provides the following implementation methods:

1) As an optional implementation, as shown in Figure 3 and Figure 4, before the above-mentioned entrance servers 102-1 and 102-2 receive the access request information to be transmitted, the data access method further includes:

S402, the ingress servers 102-1 and 102-2 acquire the delay information and packet loss rate information from themselves to other network servers in the network server set at predetermined intervals, and send the delay information and packet loss rate information to the configuration system 302 . Optionally, the network server set may include but not limited to: entry servers 102-1 and 102-2, intermediate servers 106-1 to 106-3, and destination server 104.

For example, the portal server 102-1 acquires the following information every predetermined period:

a) Delay information and packet loss rate information from the ingress server 102-1 to the intermediate server 106-1;

b) delay information and packet loss rate information from the ingress server 102-1 to the intermediate server 106-2;

c) Delay information and packet loss rate information from the entry server 102-1 to the intermediate server 106-3;

d) Delay information and packet loss rate information from the ingress server 102-1 to the destination server 104.

S404, the intermediate servers 106-1 to 106-3 and the destination server 104 obtain the delay information and packet loss rate information from themselves to other network servers in the network server set every predetermined period, and send the delay information and packet loss rate information Give configuration system 302 .

S406, the configuration system 302 calculates the link information between the ingress server 102-1 and the destination server 104 and the link information between the ingress server 102-2 and the destination server 104 according to the received delay information and packet loss rate information;

S408, the entry server 102-1 receives the link information between the entry server 102-1 and the destination server 104 returned by the configuration system 302, and the entry server 102-2 receives the link information between the entry server 102-2 and the destination server 104 returned by the configuration system 302. Link information between.

Optionally, the above S402 and S404 may be executed simultaneously, or executed sequentially.

In this embodiment, the entry servers 102-1 and 102-2 receive the link information between the entry server 102-1 and the destination server 104 returned by the configuration system 302 and the link information between the entry server 102-2 and the destination server 104. After the link information between the ingress server 102-1 and the destination server 104 is obtained from the local storage, the link information can be saved in the local storage, so that when a data access request is received information and link information between the entry server 102-2 and the destination server 104. Through the foregoing embodiments, according to dynamically allocated link information, link information can be acquired faster, access speed can be improved, and link resources can be used more reasonably.

2) As another optional method, as shown in FIG. 5 , the ingress servers 102-1 and 102-2 can also obtain delay information and packet loss rate information from themselves to other network servers under predetermined conditions, including :

S502, the ingress servers 102-1 and 102-2 receive link information acquisition requests sent by the configuration system 302, and are used to acquire delay information and packet loss rate information from themselves to other network servers in the network server set. Optionally, the network server set may include but not limited to: entry servers 102-1 and 102-2, intermediate servers 106-1 to 106-3, and destination server 104.

Optionally, the predetermined conditions include: each network server in the network server set receives the link information acquisition request sent by the configuration system 302, but this embodiment is not limited to this method, and may also include other predetermined conditions condition.

For example, when the entry server 102-1 meets the predetermined condition, it includes receiving the link information acquisition request sent by the configuration system 302: acquiring the following information:

a) Delay information and packet loss rate information from the ingress server 102-1 to the intermediate server 106-1;

b) delay information and packet loss rate information from the ingress server 102-1 to the intermediate server 106-2;

c) delay information and packet loss rate information from the entry server 102-1 to the intermediate server 106-3;

d) Delay information and packet loss rate information from the ingress server 102-1 to the destination server 104.

S504, the intermediate servers 106-1 to 106-3 and the destination server 104 receive the link information acquisition request sent by the configuration system 302, and are used to acquire the delay information and packet loss rate from itself to other network servers in the network server set information.

For example, when the intermediate server 106-1 meets the predetermined conditions, it includes receiving the link information acquisition request sent by the configuration system 302: acquiring the following information:

a) Delay information and packet loss rate information from the intermediate server 106-1 to the ingress server 102-1;

b) delay information and packet loss rate information from the intermediate server 106-1 to the intermediate server 106-2;

c) delay information and packet loss rate information from the intermediate server 106-1 to the intermediate server 106-3;

d) Delay information and packet loss rate information from the intermediate server 106-1 to the destination server 104.

In this embodiment, after receiving the link acquisition request sent by the configuration system 302, the network servers including the entry servers 102-1 and 102-2 obtain the delay information and loss information from themselves to other network servers in the network server set. Packet rate information.

Through the above-mentioned embodiment, the configuration system obtains the delay information and packet loss rate information from each network server in time from itself to other network servers in the network server set, and obtains the link information at that time from the above link data calculation, and realizes from The information of the entire link is globally perceived, which facilitates real-time data access according to requirements.

3) As yet another optional method, as shown in FIG. 6, the ingress servers 102-1 and 102-2 acquire the link information between the ingress server 102-1 and the destination server 104 and the link information between the ingress server 102-2 and the destination server. Link information between 104 includes:

S602, the entrance servers 102-1 and 102-2 transmit the first request message to the configuration system 302; optionally, the entrance servers 102-1 and 102-2 transmit the first request message in real time, and the above-mentioned first request message is used to obtain the link road information.

S604. After receiving the first transmission request, the configuration system 302 transmits a second request message to the ingress servers 102-1 and 102-2 in response to the first request. Optionally, the second request message is used to obtain delay information and packet loss rate information from the ingress servers 102-1 and 102-2 to other network servers in the network server set.

S606. After receiving the first transmission request, the configuration system 302 transmits a second request message to the intermediate servers 106-1 to 106-3 and the destination server 104 in response to the first request. Optionally, the second request message is used to obtain delay information and packet loss rate information of the intermediate servers 106-1 to 106-3 and the destination server 104 to other network servers in the network server set.

As an optional manner, after the above access request is completed, the link connection established by the entry servers 102-1 and 102-2 to the destination server 104 will also be removed.

As another optional mode, the portal server 102-1 obtains the link information between the portal server 102-1 and the destination server 104, and the portal server 102-2 obtains the link information between the portal server 102-2 and the destination server 104. After the link information, the entry servers 102-1 and 102-2 will pre-establish multiple complete link connections and put them in their corresponding connection pools. When access requests are accessed, the entry servers 102-1 and 102-2 An established connection can be taken from the connection pool for data transmission, so that the time for establishing a connection can be reduced and the effect of speeding up the transmission can be achieved; 2 The link connection established to the destination server 104 will also be removed. A new connection will be added to the connection pool immediately for subsequent access requests. In addition, all connections have a timeout period, which is determined by the destination server 104. Connections within the connection timeout period are available, and when the connection timesout, the connection will be actively removed from the connection pool, and the connection pool is always guaranteed Connection is available. Through the above embodiments, the portal server can request link information in real time when it receives a data access request from a terminal, so that the obtained link information can be updated in real time, so that link resources can be used more rationally, and the whole network can be achieved. Maximize bandwidth utilization and increase link utilization.

The scheme in which the portal server 102-1 sends the above access request information to the destination server 104 according to the above link information is described below with reference to FIGS. 7-8 and taking the portal server 102-1 as an example.

Assume that the link information indicates that the link from the ingress server 102 - 1 to the destination server 104 is: the ingress server 102 - 1 , the intermediate server 106 - 1 , the intermediate server 106 - 2 , the egress server 702 , and the destination server 104 . Based on the transmission network shown in FIG. 7, it can be seen that the ingress server 102-1 sends the above-mentioned access request information to the destination server 104 according to the above-mentioned link information, including the following steps:

S1: The ingress server 102-1 sends the above access request information to the egress server 702 connected to the destination server 104 through the intermediate servers 106-1 and 106-2 indicated by the link information; in this embodiment, the egress server 702 can view as an intermediate server.

S2: The egress server 702 sends the above access request information to the destination server 104 .

As an optional implementation of step S1, the ingress server 102-1 obtains the socket of the intermediate server 106-1 from the link information, and establishes an connect. After the connection is established successfully, the entry server 102-1 sends the link information and the access request information to the intermediate server 106-1. Then, the intermediate server 106-1 obtains the socket of the intermediate server 106-2 as the next hop of the intermediate server 106-1 from the above-mentioned link information; The intermediate server 106-2 establishes the connection.

Optionally, the above access request information includes a TCP header (TCP header) and a TCP body (TCP body), the format of which is shown in Table 1:

Table 1

TCP header TCP body

Optionally, assuming that the link is the ingress server 102-1—the intermediate server 106-1—the intermediate server 106-2—the egress server 702—the destination server 104, the link information includes the following format:

^^^Ingress server 102-1 listening socket + intermediate server 106-1 listening socket + intermediate server 106-2 socket + export server 702 socket #Destination server 104 listening socket load balancing method $ $$

For example:

^^^192.168.100.166:80+192.168.100.116:65330+192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

Optionally, when establishing a connection between servers, assemble the client IP and link information into metadata, the metadata starts with "^^^" and ends with "$$$". Use "*" after IP to distinguish, use "+" to connect between intermediate sockets, use "#" to distinguish between egress server 702 and destination server 104, "" followed by load balancing mode. Suppose the link is intermediate server 106-1——intermediate server 106-2——egress server 702——destination server 104, and the metadata format is expressed as follows:

^^^User IP*intermediate server 106-1 listening socket+intermediate server 106-2 socket+exit server 702 socket#destination server 104 listening socket load balancing method $$$

For example:

^^^192.168.100.112*192.168.100.116:65330+192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

Optionally, the entry server 102-1 inserts metadata between the TCP header and the TCP body of the access request. For example, the metadata is:

^^^192.168.100.112*192.168.100.116:65330+192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

Then the relationship between metadata and TCP header and TCP body can be shown in Table 2 below:

Table 2

If the next hop of the ingress server 102-1 is the destination server 104, then there is no need to re-encapsulate the metadata.

Optionally, if the connection between the intermediate server 106-1 and the intermediate server 106-2 is successfully established, the intermediate server 106-1 deletes the socket of the intermediate server 106-1 from the link information, and sends the updated link information and the above access request information are sent to the intermediate server 106-2. In this embodiment, the above socket may include but not limited to: IP address and port number or IP address.

For example, the link is the intermediate server 106-1—the intermediate server 106-2—the egress server 702—the destination server 104, and the format of the transmitted metadata is as follows:

^^^192.168.100.112*192.168.100.116:65330+192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

If the intermediate server 106-1 establishes a connection with the intermediate server 106-2 successfully, then modify the metadata information, remove its own socket, then forward the metadata + client IP to the backend server, and the metadata sent by the intermediate server 106-1 The format of the data information is as follows:

^^^192.168.100.112*192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

If the connection between the intermediate server 106-2 and the egress server 702 is successfully established, the format of the metadata information sent by the intermediate server 106-2 is as follows:

^^^192.168.100.112*192.168.100.191:65331#192.168.100.194:80cip$$$

Until the destination server 104, the link information will be deleted, and the reserved client IP is sent to the destination server 104,

The format is as follows Table 3:

table 3

Optionally, if the intermediate server 106-1 fails to establish a connection with the intermediate server 106-2, the intermediate server 106-1 acquires the socket of the egress server 702 as the next hop of the intermediate server 106-2 from the above link information ; The intermediate server 106-1 establishes a connection with the egress server 702 according to the socket of the egress server 702.

For example, the intermediate server 106-1 establishes a connection with the egress server 702 through the intermediate server 106-3 (as shown by the dotted line in FIG. 8 ). Optionally, if the intermediate server 106-1 fails to establish a connection with 106-3, the intermediate server 106-1 acquires the socket of the egress server 702 as the next hop of the intermediate server 106-3 from the above link information; If the connection between the intermediate server 106-1 and the export server 702 is successfully established, the intermediate server 106-1 deletes the sockets of the intermediate server 106-1 and the intermediate server 106-3 from the above-mentioned link information, and sends the updated link information and the above access request information are sent to the egress server 702. In this embodiment, the above socket may include but not limited to: IP address and port number or IP address.

For example, assuming that the link is: intermediate server 106-1——intermediate server 106-3—exit server 702——destination server 104, the format of the transmitted metadata is as follows:

^^^192.168.100.112*192.168.100.116:65330+192.168.100.190:65333+192.168.100.191:65331#192.168.100.194:80cip$$$

If the intermediate server 106-1 fails to establish a connection with the intermediate server 106-3, then skip the socket of the intermediate server 106-3, and establish a connection with the socket of the next hop socket outlet server 702, and the intermediate server 106- 1. Delete the sockets of the intermediate server 106-1 and the intermediate server 106-3 from the above link information, and send the updated link information and the above access request information to the egress server 702. The format of the transmitted metadata is as follows :

^^^192.168.100.112*192.168.100.191:65331#192.168.100.194:80cip$$$

After the egress server 702 receives the updated link information and the above-mentioned access request information, it obtains the socket of the destination server 104 as the next hop of the egress server 702 from the above-mentioned link information; The socket establishes a connection with the destination server 104 ; if the connection between the egress server 702 and the destination server 104 is successfully established, the egress server 702 sends the access request information to the destination server 104 . That is to say, the egress server 702 will not send the link information for network transmission to the destination server 104 .

Through the above-mentioned embodiment, the socket used to establish the connection is obtained by parsing the dynamically allocated link information without pre-configuration, thereby reducing the complexity of configuration and speeding up the data transmission speed; further, The IP of the entry server can be brought to the destination server, so that the destination server can obtain the IP of the netizen.

Example 2

This embodiment provides an entry server, taking the entry server 102-1 as an example, as shown in FIG. 9, the entry server 102-1 in this embodiment includes:

1) The first transmission unit 802: used to receive the access request information to be transmitted, wherein the above access request information is used to access data on the destination server;

2) Obtaining unit 804: used to obtain link information between the above-mentioned entry server and the above-mentioned destination server, wherein the link information is reported by the configuration system according to the network server including the entry server and the destination server from itself to the set of network servers Delay information and packet loss rate information of other network servers in the network are calculated;

3) The second transmission unit 806: configured to send the above access request information to the above destination server according to the above link information.

As an optional manner, taking the entry server 102-1 as an example, after the above access request is completed, the link connection established by the entry server 102-1 to the destination server 104 will also be removed.

As another optional way, taking the ingress server 102-1 as an example, after the ingress server 102-1 obtains the link information between the ingress server 102-1 and the destination server 104, the ingress server 102-1 will pre-establish Multiple complete links are connected and placed in the connection pool. When an access request is accessed, the entry server 102-1 can take an established connection from the connection pool for data transmission. In this way, the time for establishing a connection can be reduced. time, which has the effect of speeding up the transmission; when the above-mentioned access request is completed, the link connection established by the entry server 102-1 to the destination server 104 will also be removed. A new connection will be added to the connection pool immediately for subsequent access requests. In addition, all connections have a timeout period, which is determined by the destination server 104. Connections within the connection timeout period are available, and when the connection timesout, the connection will be actively removed from the connection pool, and the connection pool is always guaranteed Connection is available.

In this embodiment of the present invention, the entry server 102-1 receives access request information for accessing data on the destination server 104 through the first transmission unit 802, and the acquisition unit 804 obtains the above-mentioned entry dynamically assigned by the configuration system 302 The above link information between the server 102-1 and the above destination server 104, and then, the second transmission unit 806 sends the above access request information to the above destination server 104 according to the above link information. Optionally, the above-mentioned link information may include but not limited to: the intermediate server 106-1, the intermediate server 106-2, and the socket of the intermediate server 106-2. Here, the socket includes: an IP address and a port number or an IP address.

Through the above embodiments, the link information dynamically allocated by the configuration system can reduce configuration complexity and speed up the transmission of dynamic content.

As an optional manner, still taking the entry server 102-1 as an example, as shown in FIG. 10, the above entry server 102-1 also includes:

1) The third transmission unit 902 is configured to acquire the delay information and Packet loss rate information, wherein the above-mentioned network server set includes the above-mentioned entrance server and the above-mentioned destination server; sending the above-mentioned delay information and the above-mentioned packet loss rate information to the above-mentioned configuration system;

2) The fourth transmission unit 904 is configured to receive the above-mentioned link information between the above-mentioned ingress server and the above-mentioned destination server returned by the above-mentioned configuration system, wherein the above-mentioned link information is provided by the above-mentioned configuration system according to the above-mentioned delay information and the above-mentioned packet loss Rate information is calculated;

Wherein, the obtaining unit is configured to locally obtain the link information between the entry server and the destination server.

In the embodiment of the present invention, before receiving the above-mentioned access request information to be transmitted, the third transmission unit 902 of the portal server 102-1 obtains the above-mentioned portal server 102-1 to The delay information and packet loss rate information of the intermediate servers 106-1 to 106-3 or the destination server 104; the obtained delay information and packet loss rate information are sent to the configuration system 302, and the fourth transmission unit 904 receives the configuration system 302 calculates the link information according to the above data, and sends it to the entry server 102-1.

As an optional manner, the predetermined condition includes: the network server receives the link information acquisition request sent by the configuration system 302 .

In the embodiment of the present invention, the ingress server 102-1 may also send the above delay information and packet loss rate information to the configuration system 302 after receiving the link information request sent by the configuration system 302, in response to the request.

Through the above-mentioned embodiments, during the data access process, the information of the entire link can be sensed in advance from the global perspective or with predetermined conditions, and link resources can be used more rationally by cooperating with the link detection and link calculation server, so as to achieve full network Maximize bandwidth utilization and increase link utilization.

As an optional manner, as shown in FIG. 11, the acquisition unit 804 includes:

1) The first transmission module 1002, configured for the above-mentioned configuration system to send a first request message for obtaining the above-mentioned link information;

2) A second transmission module 1004, configured to receive a second request message for acquiring link information sent by the configuration system in response to the first request message;

3) The third transmission module 1006, configured to send delay information and packet loss rate information from the above-mentioned entrance server to other network servers in the network server set to the above-mentioned configuration system in response to the above-mentioned second request message, wherein the above-mentioned network server The collection includes the above-mentioned entry server and the above-mentioned destination server;

4) The fourth transmission module 1008 is configured to receive the above-mentioned link information between the above-mentioned ingress server and the above-mentioned destination server and calculated according to the above-mentioned delay information and the above-mentioned packet loss rate information sent by the above-mentioned configuration system.

As an optional manner, taking the entry server 102-1 as an example, after the above access request is completed, the link connection established by the entry server 102-1 to the destination server 104 will also be removed.

As another optional way, taking the ingress server 102-1 as an example, after the ingress server 102-1 obtains the link information between the ingress server 102-1 and the destination server 104, the ingress server 102-1 will pre-establish Multiple complete links are connected and placed in the connection pool. When an access request is accessed, the entry server 102-1 can take an established connection from the connection pool for data transmission. In this way, the time for establishing a connection can be reduced. time, which has the effect of speeding up the transmission; when the above-mentioned access request is completed, the link connection established by the entry server 102-1 to the destination server 104 will also be removed. A new connection will be added to the connection pool immediately for subsequent access requests. In addition, all connections have a timeout period, which is determined by the destination server 104. Connections within the connection timeout period are available, and when the connection timesout, the connection will be actively removed from the connection pool, and the connection pool is always guaranteed Connection is available.

In this embodiment of the present invention, the entry server 102-1 sends a first request message to the configuration system 302 by the first transmission module 1002 to obtain link information; the entry server 102-1 receives the above-mentioned The second request message sent by the configuration system 302 in response to the above-mentioned first request message for obtaining link information is used to obtain link information; the third transmission module 1006 of the ingress server 102-1 responds to the second request message to the above-mentioned The configuration system 302 sends itself to the delay information and packet loss rate information of other network servers in the network server set including the first intermediate server 106-1 to 106-3 or the destination server 104; Link information is calculated from the path data, and sent to the ingress server 102-1, and received by the fourth transmission module 1008.

Optionally, the above access request information includes a TCP header (TCP header) and a TCP body (TCP body), the format of which is shown in Table 4:

Table 4

TCP header TCP body

Optionally, assuming that the link is the ingress server 102-1—the intermediate server 106-1—the intermediate server 106-2—the egress server 702—the destination server 104, the link information includes the following format:

^^^Ingress server 102-1 listening socket + intermediate server 106-1 listening socket + intermediate server 106-2 socket + export server 702 socket #Destination server 104 listening socket load balancing method $ $$

For example:

^^^192.168.100.166:80+192.168.100.116:65330+192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

Optionally, when establishing a connection between servers, assemble the client IP and link information into metadata, the metadata starts with "^^^" and ends with "$$$". Use "*" after IP to distinguish, use "+" to connect between intermediate sockets, use "#" to distinguish between egress server 702 and destination server 104, "" followed by load balancing mode. Suppose the link is intermediate server 106-1——intermediate server 106-2——egress server 702——destination server 104, and the metadata format is expressed as follows:

^^^User IP*intermediate server 106-1 listening socket+intermediate server 106-2 socket+exit server 702 socket#destination server 104 listening socket load balancing method $$$

For example:

^^^192.168.100.112*192.168.100.116:65330+192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

Optionally, the entry server 102-1 inserts metadata between the TCP header and the TCP body of the access request. For example, the metadata is:

^^^192.168.100.112*192.168.100.116:65330+192.168.100.180:65332+192.168.100.191:65331#192.168.100.194:80cip$$$

The relationship between metadata and TCP header and TCP body can be shown in Table 5 below:

table 5

If the next hop of the ingress server 102-1 is the destination server 104, then there is no need to re-encapsulate the metadata.

Through the above-mentioned embodiment, the information of the entire link is perceived globally, and the link is calculated according to the acquired link information perceived by each network server, so that link resources can be used more reasonably, and the bandwidth utilization of the entire network can be maximized. Increase link utilization; in addition, the socket used to establish a connection is parsed from dynamically allocated link information without pre-configuration, thereby reducing the complexity of configuration and speeding up data transmission ; Further, the IP of the entry server can be brought to the destination server, so that the destination server can obtain the IP of the netizen.

Obviously, those skilled in the art should understand that each module or each step of the above-mentioned application can be realized by a general-purpose computing device, and they can be concentrated on a single computing device, or distributed in a network composed of multiple computing devices Alternatively, they may be implemented in program code executable by a computing device so that they may be stored in a storage device to be executed by a computing device, and in some cases in an order different from that shown here The steps shown or described are carried out, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps among them are fabricated into a single integrated circuit module for implementation. As such, the present application is not limited to any specific combination of hardware and software.

The above descriptions are only preferred embodiments of the present application, and are not intended to limit the present application. For those skilled in the art, there may be various modifications and changes in the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims (10)

1. A data access method, characterized in that, comprising: The entry server receives the access request information to be transmitted, wherein the access request information is used to access data on the destination server; The ingress server acquires link information between the ingress server and the destination server, wherein the link information is reported by the configuration system according to the network server including the ingress server and the destination server. Delay information and packet loss rate information to other network servers in the network server set are calculated; The entry server sends the access request information to the destination server according to the link information. 2. The method of claim 1, wherein, Before the ingress server receives the access request information to be transmitted, the method further includes: Each network server in the set of network servers including the entry server and the destination server obtains all information from itself to other network servers in the set of network servers every predetermined period or when a predetermined condition is met. The delay information and the packet loss rate information; The network server sends the delay information and the packet loss rate information to the configuration system; The ingress server receives the link information between the ingress server and the destination server returned by the configuration system; The acquiring, by the ingress server, link information between the ingress server and the destination server includes: acquiring, by the ingress server, the link information between the ingress server and the destination server locally. 3. The method according to claim 2, wherein the predetermined condition comprises: the network server receives a link information acquisition request sent by the configuration system. 4. The method according to claim 1, wherein the acquisition by the ingress server of link information between the ingress server and the destination server comprises: The configuration system receives a first request message sent by the ingress server for obtaining the link information; The configuration system sends a second request message for obtaining link information to each network server in the network server set including the entry server and the destination server in response to the first request message; The configuration system receives delay information and packet loss rate information sent by the network server from itself to other network servers in the network server set; The configuration system calculates the link information between the ingress server and the destination server according to the delay information and the packet loss rate information; The configuration system sends the link information between the entry server and the destination server to the entry server. 5. The method according to claim 1, wherein the ingress server sending the access request information to the destination server according to the link information comprises: The ingress server sends the access request information to the egress server connected to the destination server through multiple intermediate servers indicated by the link information; The egress server sends the access request information to the destination server; Wherein, transmitting the access request information between the plurality of intermediate servers includes the following steps: The first intermediate server among the plurality of intermediate servers acquires the socket of the second intermediate server as the next hop of the first intermediate server from the link information; The first intermediate server establishes a connection with the second intermediate server according to the socket of the second intermediate server; If the first intermediate server successfully establishes the connection with the second intermediate server, the first intermediate server deletes the socket of the first intermediate server from the link information, and updates the sending the link information and the access request information to the second intermediate server; Wherein, the socket is an IP address and a port number or is an IP address. 6. The method according to claim 5, wherein transmitting the access request information between the multiple intermediate servers comprises: If the first intermediate server fails to establish the connection with the second intermediate server, the first intermediate server obtains the set of the third intermediate server that is the next hop of the second intermediate server from the link information. Receive words; The first intermediate server establishes a connection with the third intermediate server according to the socket of the third intermediate server; If the first intermediate server successfully establishes the connection with the third intermediate server, the first intermediate server deletes all links between the first intermediate server and the second intermediate server from the link information. the socket, and send the updated link information and the access request information to the third intermediate server. 7. An entry server, characterized in that, comprising: A first transmission unit, configured to receive access request information to be transmitted, wherein the access request information is used to access data on the destination server; An acquisition unit, configured to acquire link information between the entry server and the destination server, wherein the link information is reported by the configuration system according to the network server including the entry server and the destination server Delay information and packet loss rate information from itself to other network servers in the network server set are calculated; A second transmission unit, configured to send the access request information to the destination server according to the link information. 8. The ingress server according to claim 7, comprising: A third transmission unit, configured to obtain the delay information from the entry server to other network servers in the network server set at a predetermined period or when a predetermined condition is met before receiving the access request information to be transmitted and the packet loss rate information, wherein the set of network servers includes the entry server and the destination server; sending the delay information and the packet loss rate information to the configuration system; A fourth transmission unit, configured to receive the link information between the entry server and the destination server returned by the configuration system; Wherein, the obtaining unit is configured to locally obtain the link information between the entry server and the destination server. 9. The portal server according to claim 8, wherein the predetermined condition comprises: the network server receives a link information acquisition request sent by the configuration system. 10. The portal server according to claim 7, wherein the acquiring unit comprises: A first transmission module, configured for the configuration system to send a first request message for acquiring the link information; A second transmission module, configured to receive a second request message for acquiring link information sent by the configuration system in response to the first request message; A third transmission module, configured to send delay information and packet loss rate information from the ingress server to other network servers in the network server set to the configuration system in response to the second request message, wherein the network The set of servers includes the entry server and the destination server; The fourth transmission module is configured to receive the link information between the ingress server and the destination server and calculated according to the delay information and the packet loss rate information sent by the configuration system.

Images