WO2017167129A1 - 一种为用户分配数据中心的方法和设备 - Google Patents

一种为用户分配数据中心的方法和设备 Download PDF

Info

Publication number
WO2017167129A1
WO2017167129A1 PCT/CN2017/078044 CN2017078044W WO2017167129A1 WO 2017167129 A1 WO2017167129 A1 WO 2017167129A1 CN 2017078044 W CN2017078044 W CN 2017078044W WO 2017167129 A1 WO2017167129 A1 WO 2017167129A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
data center
traffic
data
idc
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2017/078044
Other languages
English (en)
French (fr)
Inventor
李彦超
陈禹
桑植
郭东白
邹廷明
温凯
吴倩倩
周志伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Publication of WO2017167129A1 publication Critical patent/WO2017167129A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/40—Network security protocols
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/12—Avoiding congestion; Recovering from congestion
    • H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1031—Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/563—Data redirection of data network streams

Definitions

  • the present disclosure relates to the field of network communication technologies, and in particular, to a method and device for allocating a data center to a user.
  • the Internet needs to provide multiple data centers in order to better meet the needs of users. Since the access speed of the user is often different depending on the location of the data center, the closer the data center to which the user belongs is to the user, the faster the access speed. Therefore, when assigning a user's home data center, the user needs to be assigned to the data center closest to it.
  • Internet companies tend to face unexpected bursts of traffic growth. In the face of sudden traffic growth, because different data centers can carry user-generated traffic is different, and some data centers cannot carry users with traffic near the peak. Traffic, which requires a reasonable allocation of users attributable to the data center as burst traffic grows. For example, Facebook Group's annual global double eleven shopping carnival. If the data center is allocated to the user by the method of assigning the data center nearby, if most of the carnival users are concentrated in a certain area, the problem of excessive local traffic may occur.
  • the domain name is resolved based on the geographical area to which the address segment of the user's IP (Internet Protocol) is based on the DNS (Domain Name System) technology.
  • IP Internet Protocol
  • DNS Domain Name System
  • the nearest computer room thus allocating data centers to users.
  • multiple users may share the same IP address, and one user may use multiple IP addresses.
  • Assign data to users based on IP address segments In the central method, when the traffic peak period, the actual number of users in each data center cannot be accurately determined, and the data center cannot be allocated to the user reasonably.
  • the current method of assigning a data center to a user based on an IP address segment cannot reasonably allocate a data center to a user during a peak traffic period.
  • the present disclosure provides a method and a device for allocating a data center to a user, which is used to solve the problem that a data center is allocated to a user based on an IP address segment in the prior art, and the data center cannot be reasonably allocated to the user during a peak traffic period. .
  • an embodiment of the present disclosure provides a method for allocating a data center to a user, including:
  • the first data center is a data center whose peak traffic ratio is greater than the proportion of the allocated traffic
  • the second data center is a data center whose peak traffic ratio is smaller than the proportion of the allocated traffic
  • the embodiment of the present disclosure uses the data center whose peak traffic ratio is greater than the proportion of the allocated traffic as the first data center, and determines the users that need to be transferred in the first data center, and the data that accounts for a larger proportion of the distributed traffic than the peak traffic.
  • the center performs a transfer process on the users in the first data center that need to be transferred according to the second data center, to ensure that each data center can satisfy all users belonging to the data center.
  • the proportion of the allocated traffic and the peak traffic of each data center when there is a difference between the proportion of the allocated traffic of the data center and the peak traffic, the data center to which the user belongs is adjusted, which is reasonable for the user. Assign a data center.
  • the ratio of the allocated traffic of the data center is a ratio of an upper limit of the data traffic carried by the data center to an upper limit of the upper limit of user traffic carried by the data center.
  • the determining the peak traffic proportion of each data center includes:
  • the ratio of the sum of the traffic weights of the users belonging to the data center to the sum of the traffic weights of all users belonging to all data centers is used as the peak traffic of the data center. Proportion.
  • the embodiment of the present disclosure provides a method for determining the peak traffic ratio of the data center according to the traffic weight, the peak traffic proportion of the data center is accurately calculated.
  • the traffic weight of the user is a value indicating a proportion of user traffic generated by the user to a sum of user traffic generated by all users of all data centers.
  • the peak traffic ratio of the data center can be obtained according to the traffic weight of the user.
  • the process of transferring the user that needs to be transferred in the first data center to the determined second data center includes:
  • the embodiment of the present disclosure is directed to a first data center, the user who needs to transfer is selected from the first data center, and the user who needs to transfer in the first data center is transferred to the second data center to ensure each data center. Both can be used by all users belonging to the data center.
  • the determining, in the first data center, that the user needs to perform the transfer includes:
  • the embodiment of the present disclosure determines the amount of traffic that needs to be transferred in the first data center according to the proportion of traffic of each user in the first data center, the sum of the traffic ratios is selected from the first data center.
  • the user who is not less than the difference between the peak traffic proportion and the allocated traffic ratio of the first data center enables the first data center to satisfy the user belonging to the first data center, thereby appropriately allocating the data center to the user.
  • the user that needs to be transferred from all the users that belong to the first data center according to the ratio of the user traffic includes:
  • the top N users ranked from high to low are used as the users who need to transfer;
  • N is a positive integer.
  • the N users with the largest user traffic ratio are selected as the need to transfer.
  • the user, and the sum of the user traffic ratios of the selected N users is not less than the difference between the peak traffic proportion of the first data center and the proportion of the allocated traffic, thereby implementing data in the case of influencing the least number of users.
  • Reasonable distribution of the center when determining the user that needs to be transferred in the first data center, according to the proportion of user traffic of all users in the first data center, the N users with the largest user traffic ratio are selected as the need to transfer.
  • the user, and the sum of the user traffic ratios of the selected N users is not less than the difference between the peak traffic proportion of the first data center and the proportion of the allocated traffic, thereby implementing data in the case of influencing the least number of users.
  • determining, from the second data center that is capable of receiving the user, the at least one of the second data centers that can receive the user in the first data center that needs to be transferred including:
  • the transfer principle includes one of the following transfer principles:
  • the second data center that matches the first data center Determining, for any one of the first data centers, the second data center that matches the first data center, wherein all of the second data centers that match the first data center are sorted
  • the location in the second data center is not lower than the location of the second data center that matches the target first data center in all of the sorted second data centers, the target first data center is sorted a location in all of the first data centers is lower than a location of the first data center in all of the sorted first data centers;
  • the first difference of the first data center is a difference between a peak traffic proportion of the first data center and an allocated traffic ratio of the first data center; and a second of the second data center The difference is the difference between the ratio of the allocated traffic of the second data center and the proportion of the peak traffic of the second data center;
  • the embodiment of the present disclosure provides a plurality of transfer rules for determining a user who needs to transfer in the first data center, for determining a second data center of the user who needs to transfer the first data center, and a different transfer method, This enables a flexible allocation of data centers for users.
  • the binding relationship between the data center and the user is generated according to the following manner:
  • the embodiment of the present disclosure uniquely identifies the user according to the user ID, and determines the attribution of the user's long-term login, the data center closest to the home location where the user has long-logged in is used as the data of the user's attribution.
  • the center avoids the inaccuracy of determining the data center to which the user belongs based on the IP address, thereby accurately assigning the user to the nearest data center.
  • the step of determining the first data center that needs to perform user transfer and the second data center capable of receiving the user is completed during a busy time period;
  • the step of determining the attribution of the user multiple times according to the access log of the user at different times is completed in a non-busy time period.
  • the embodiments of the present disclosure provide a method of separately assigning data centers to users for busy periods and non-busy periods.
  • the data center is allocated to the user; in the non-busy period, according to the proportion of the allocated traffic and the peak traffic of each data center, the users belonging to the data center are reasonably adjusted, thereby flexibly and accurately assigning data to the user. center.
  • the method further includes:
  • the data center corresponding to the home location when the user logs in is determined to be the same one, and the user logs in.
  • the data center corresponding to the home is different from the data center to which the user is currently bound, and the data center corresponding to the home address when the user logs in is used as the data center bound by the user.
  • the embodiment of the present disclosure determines the changed attribution after the change of the attribution of the user, if the attribution after the change in the set duration is the same, it is determined that the attribution of the user changes, and the modification is performed.
  • the user is currently bound to the data center to better satisfy the user's access to the Internet through the nearest data center.
  • the method further includes:
  • the user data whose data center belongs to the user needs to be transferred, and the user data of the user is changed from the original.
  • the data center to which it belongs is migrated to the changed data center.
  • the method further includes:
  • the currently bound data center is controlled to provide network services for the user.
  • an embodiment of the present disclosure further provides an apparatus for allocating a data center to a user, including:
  • An obtaining module configured to determine an allocated traffic ratio of each data center, and determine a peak traffic proportion of each of the data centers
  • a determining module configured to determine a first data center that needs to perform user transfer and a second data center that can accept the user
  • a processing module configured to transfer, to the determined second data center, a user that needs to be transferred in the first data center, where the first data center is greater than the proportion of the allocated traffic
  • the ratio of the allocated traffic of the data center is a ratio of an upper limit of the data center carrying user traffic to an upper limit of the data center carrying user traffic.
  • the acquiring module is specifically configured to:
  • the ratio of the sum of the traffic weights of the users belonging to the data center to the sum of the traffic weights of all users belonging to all data centers is used as the peak traffic of the data center. Proportion.
  • processing module is specifically configured to:
  • processing module is specifically configured to:
  • Determining a proportion of user traffic belonging to each user of the first data center selecting, according to the user traffic ratio, all users that need to be transferred from all the users belonging to the first data center, wherein the selected ones are selected
  • the sum of the user traffic proportions of the user is not less than a difference between the peak traffic proportion of the first data center and the allocated traffic ratio.
  • processing module is specifically configured to:
  • the top N users ranked from high to low are used as the users who need to transfer; wherein N is a positive integer.
  • processing module is specifically configured to:
  • the transfer principle includes one of the following transfer principles:
  • Transfer principle 1 Sort all the first data centers according to the first difference of all the first data centers, from large to small; and according to the second difference of all the second data centers, All of the second data centers are sorted from largest to smallest;
  • the second data center that matches the first data center Determining, for any one of the first data centers, the second data center that matches the first data center, wherein all of the second data centers that match the first data center are sorted
  • the location in the second data center is not lower than the location of the second data center that matches the target first data center in all of the sorted second data centers, the target first data center is The positions in all of the sorted first data centers are lower than the positions of the first data centers in all of the sorted first data centers;
  • the first difference of the first data center is a difference between a peak traffic proportion of the first data center and an allocated traffic ratio of the first data center; and a second of the second data center The difference is the difference between the ratio of the allocated traffic of the second data center and the proportion of the peak traffic of the second data center;
  • processing module is further configured to generate the binding relationship between the data center and the user according to the following manner:
  • the processing module is specifically configured to: the step of determining the first data center that needs to perform user transfer and the second data center capable of receiving the user is performed in a busy time period;
  • the step of determining the attribution of the user multiple times according to the access log of the user at different times is completed in a non-busy time period.
  • processing module is further configured to:
  • the data center corresponding to the home location when the user logs in is determined to be the same one, and the user logs in.
  • the data center corresponding to the home is different from the data center to which the user is currently bound, and the data center corresponding to the home address when the user logs in is used as the data center bound by the user.
  • processing module is further configured to:
  • processing module is further configured to:
  • the currently bound data center is controlled to provide network services for the user.
  • FIG. 1A is a schematic diagram 1 of a mapping relationship between a user and an IDC according to an embodiment of the present disclosure
  • FIG. 1B is a schematic diagram of a routing table 1 according to an embodiment of the present disclosure.
  • FIG. 2 is a flowchart of a method for allocating an IDC to a user according to an embodiment of the present disclosure
  • FIG. 3 is a second schematic diagram of a mapping relationship between a user and an IDC according to an embodiment of the present disclosure
  • FIG. 4 is a schematic diagram 3 of a mapping relationship between a user and an IDC according to an embodiment of the present disclosure
  • FIG. 5 is a schematic diagram of a routing table 2 according to an embodiment of the present disclosure.
  • 6A is a schematic diagram of mapping relationship between a user and an IDC before a user transfer according to an embodiment of the present disclosure
  • 6B is a schematic diagram of a pre-transfer routing table of the embodiment of the present disclosure.
  • 6C is a schematic diagram of a mapping relationship between a user and an IDC after a user transfer according to an embodiment of the present disclosure
  • 6D is a schematic diagram of a user-transferred routing table according to an embodiment of the present disclosure.
  • FIG. 7 is an overall flowchart of a method for allocating an IDC to a user according to an embodiment of the present disclosure
  • FIG. 8 is a flowchart of a method for generating a binding relationship between an IDC and a user of a home IDC according to an embodiment of the present disclosure
  • FIG. 9 is a flowchart of a method for an IDC to provide a service to a user according to an embodiment of the present disclosure
  • FIG. 10 is a schematic structural diagram of an apparatus for allocating a data center to a user according to an embodiment of the present disclosure.
  • Embodiments of the present disclosure determine an allocation traffic ratio for each data center, and determine a peak traffic ratio for each of the data centers; determine a first data center that requires user transfer and a second data center that can accept users; Transmitting, to the determined second data center, the user that needs to be transferred in the first data center; wherein the first data center is a data center whose peak traffic ratio is greater than the proportion of the allocated traffic, The data center is a data center in which the peak traffic ratio is smaller than the proportion of the allocated traffic.
  • the embodiment of the present disclosure uses the data center whose peak traffic ratio is greater than the proportion of the allocated traffic as the first data center, and determines the users that need to be transferred in the first data center, and the data that accounts for a larger proportion of the distributed traffic than the peak traffic.
  • the center serves as a second data center. According to the second data center, the users in the first data center that need to be transferred are transferred to ensure that each data center can satisfy all users belonging to the data center.
  • the proportion of the allocated traffic and the peak traffic of each data center when there is a difference between the proportion of the allocated traffic of the data center and the peak traffic, the data center to which the user belongs is adjusted, which is reasonable for the user. Assign a data center.
  • Embodiments of the present disclosure may be applied to scenarios where multiple data centers provide services to users.
  • the data center needs to be allocated to the user according to multiple data centers to improve the user experience.
  • the data center of the embodiment of the present disclosure may be an IDC (Internet Data Center), and may be another center for storing data.
  • IDC Internet Data Center
  • the processing method of the data center is another center for storing data is the same as that of the data center, and will not be described here.
  • the IDC of the embodiment of the present disclosure is specifically an Internet data center, in which a certain number of network servers are included, and specific Internet software is deployed on the network server to provide network services to users of the Internet.
  • mapping relationship of the user to its nearest IDC which may also be referred to as a routing table.
  • the user's mapping relationship to its nearest IDC is set
  • the elements in the set are the mapping relationship between the user and the IDC to which the user belongs, such as ⁇ user_id, IDC>.
  • the set of mappings between all users and the IDC to which the user belongs is a routing table, and the routing table can be represented by Set ⁇ user_id, IDC>>.
  • the IDC of the embodiment of the present disclosure includes IDC1, IDC2, and IDC3, and the user includes User A, User B, User C, User D, User E, User F, User G, and User H.
  • A, B, C, D, E, F, G, and H are user IDs, and the mapping relationship between the user and the IDC is as shown in FIG. 1A.
  • User A, user B, and user E belong to IDC1; user C and user F And user G belongs to IDC2; user D and user H belong to IDC3.
  • the set of the mapping relationship between all users and the IDC to which the user belongs is represented by a routing table.
  • the routing table is as shown in FIG.
  • a method for allocating an IDC to a user includes:
  • Step 201 Determine an allocation ratio of each IDC, and determine a peak traffic proportion of each IDC.
  • Step 202 Determine a first IDC that needs to perform user transfer and a second IDC that can accept the user;
  • Step 203 Transfer the user that needs to be transferred in the first IDC to the determined second IDC, where the first IDC is an IDC whose peak traffic ratio is greater than the proportion of the allocated traffic.
  • the second IDC is an IDC whose peak traffic ratio is smaller than the proportion of the allocated traffic.
  • the embodiment of the present disclosure assigns a unique identification ID to each user on the Internet, and the embodiment of the present disclosure identifies the user according to the user ID. For example, for users of Facebook.com, when users use Facebook.com, they will be asked to register on their website. When the user is successfully registered, the user ID will be assigned to the user.
  • the ratio of the allocated traffic of the data center of the embodiment of the present disclosure is the ratio of the upper limit of the data center carrying user traffic to the upper limit of the total data center carrying user traffic.
  • the proportion of the allocated traffic of the data center of the embodiment of the present disclosure is pre-configured.
  • an assigned traffic ratio is pre-allocated for each IDC.
  • ⁇ IDC1, 30%> means that 30% of the traffic is pre-owned in IDC1, that is, IDC1 carries up to 30% of the global traffic.
  • the embodiment of the present disclosure can directly obtain the proportion of the allocated traffic pre-allocated for each IDC.
  • the embodiment of the present disclosure pre-configures the allocation traffic ratio, it is determined according to the number of network servers corresponding to each IDC. Specifically, for any IDC, when determining the proportion of the allocated traffic of the IDC, the ratio of the number of network servers corresponding to the IDC to the number of all network servers corresponding to all the IDCs is taken as the proportion of the allocated traffic of the IDC.
  • the embodiment of the present disclosure includes three IDCs, namely IDC1, IDC2, and IDC3. There are 30 network servers corresponding to IDC1, 30 network servers corresponding to IDC2, and 40 network servers corresponding to IDC3. IDC1 and IDC2 account for 30% of the allocated traffic, and IDC3's allocated traffic accounts for 40%.
  • the peak traffic ratio of the data center of the embodiment of the present disclosure is the ratio of the sum of the traffic weights of the users belonging to the data center to the sum of the traffic weights of all users belonging to all data centers.
  • the weight value indicates the ratio of the sum of user traffic that may be generated by the user corresponding to the data center to the sum of user traffic that may be generated by all users belonging to all data centers, as the peak traffic proportion of the data center.
  • the peak traffic ratio of each IDC of the embodiment of the present disclosure is calculated according to the following steps.
  • All users of the embodiments of the present disclosure refer to users who use the Internet, specifically, users who have registered on the Internet, and registered users have a user unique identification ID on the Internet.
  • users who have registered on the Internet and registered users have a user unique identification ID on the Internet.
  • registered users have a user unique identification ID on the Internet.
  • the embodiment of the present disclosure predicts the traffic weight value that the user may generate through the machine learning algorithm.
  • the traffic weight value of the user in the embodiment of the present disclosure is a value indicating a proportion of user traffic generated by the user to a sum of user traffic generated by all users of all data centers.
  • the user's traffic weight is a score of 0-100 minutes, which is used to identify the traffic weight that the user can generate.
  • the embodiment of the disclosure first determines the ratio of the user traffic that may be generated by the user to the sum of the user traffic generated by the global user, and determines the traffic of the user according to the correspondence between the traffic ratio and the traffic weight. Weight.
  • the ratio of the user traffic that may be generated by user A to the user traffic generated by the global user is 0.02, and the corresponding traffic weight is assumed to be between 0.015 and 0.02. If it is 80, it is determined that the traffic of user A is 80.
  • the specific algorithm type of the machine algorithm used to determine the traffic weight of the user is not limited, and any method capable of predicting the traffic weight that the user may generate according to the historical behavior data of the user is applicable to the present disclosure. content.
  • the ratio of the sum of the traffic weights of the users belonging to the IDC to the sum of the traffic weights of all users belonging to all IDCs is taken as the peak traffic ratio of the IDC.
  • any IDC all users belonging to the IDC are determined, and the traffic weights of all users belonging to the IDC are added to obtain the traffic weight of the IDC; Add, the global traffic weight is obtained, and the ratio of the traffic weight of the IDC to the global traffic weight is taken as the peak traffic proportion of the IDC.
  • the peak traffic ratio of each IDC is the maximum traffic ratio that each IDC may generate. In an actual process, the traffic ratio generated by each IDC is less than or equal to the peak traffic ratio.
  • the global ID includes three IDCs, namely IDC1, IDC2, and IDC3; the global users include user A, user B, user C, user D, user E, user F, and user G, where the mapping relationship between the user and the IDC
  • users belonging to IDC1 have user A and user B; users belonging to IDC2 have user C, user D, and user E; users belonging to IDC3 have user F and user G.
  • the traffic weight of user A is 60
  • the traffic weight of user B is 70
  • the traffic weight of user C is 55
  • the traffic weight of user D is 40
  • the traffic weight of user E is 80
  • the traffic of user F is
  • the weight is 65
  • the traffic weight of user G is 90.
  • All IDCs of the present disclosure include, but are not limited to, the first IDC and the second IDC.
  • the first IDC is an IDC that determines that the peak traffic ratio is greater than the proportion of the allocated traffic according to the proportion of the allocated traffic and the peak traffic. That is, during the busy period, the traffic that the first IDC can provide cannot satisfy the user that belongs to the IDC.
  • the traffic to be used, the first IDC is the IDC that needs to be transferred by the user.
  • the second IDC is an IDC that determines that the peak traffic ratio is smaller than the proportion of the allocated traffic according to the proportion of the allocated traffic and the peak traffic. That is, during the busy period, the second IDC can provide more traffic than the user belonging to the IDC needs to use. Traffic, the second IDC is an IDC capable of accepting users.
  • the global user includes user A, user B, user C, user D, and user F.
  • the mapping relationship between the user and the IDC to which the user belongs is as shown in FIG. 4, and the user belonging to IDC1 has user A, user B, and user E.
  • the user belonging to IDC2 has user C, user D and user F; and the traffic weight of user A is 60, the traffic weight of user B is 70, the traffic weight of user C is 55, and the traffic weight of user D is 40, user E has a traffic weight of 80, and user F has a traffic weight of 65.
  • the peak traffic ratio of IDC1 is greater than the proportion of the allocated traffic, and then IDC1 is determined to be the first IDC, that is, some users of IDC1 need to be transferred; IDC2 is determined. If the peak traffic ratio is less than the allocated traffic ratio, it is determined that IDC2 is the second IDC, that is, IDC2 can accept the users in IDC1 that need to be transferred.
  • the embodiment of the present disclosure After determining the first IDC that needs to perform user transfer and the second IDC capable of accepting the user, the embodiment of the present disclosure also needs to determine the user in the first IDC that needs to be transferred. Since the method for determining the user who needs to transfer in the first IDC is the same for each first IDC, a first IDC is taken as an example for description.
  • the following may be specifically adopted in the following manner:
  • the embodiment of the present disclosure determines, for any one of the first IDCs, a user in the first IDC that needs to be transferred according to the following steps.
  • Embodiments of the present disclosure may determine all users belonging to the first IDC according to a routing table.
  • the routing table of the embodiment of the present disclosure is as shown in FIG. 5, namely: Set ⁇ User A, IDC1>, ⁇ User B, IDC1>, ⁇ User C, IDC2>, ⁇ User D, IDC2>, ⁇ User E, IDC2>, ⁇ user F, IDC3>>, and if IDC2 is the first IDC, when determining all users corresponding to the first IDC, according to the routing table, all users of the home IDC2 are determined to be user D, user E, and User F.
  • determining a proportion of user traffic belonging to each user of the first IDC selecting, according to the user traffic ratio, a user who needs to be transferred from all users belonging to the first IDC, where the selected location
  • the sum of the user traffic ratios of the users is not less than the difference between the peak traffic ratio of the first IDC and the proportion of the allocated traffic.
  • the user traffic ratio of the user in the embodiment of the present disclosure is the ratio of the user's traffic weight to the sum of all global user traffic weights.
  • all users in the global network include user A, user B, user C, user D, user E, and user F, and the traffic weight of user A is 60, the traffic weight of user B is 70, and the traffic weight of user C is 50.
  • the traffic weight of user D is 40, the traffic weight of user E is 80, and the traffic weight of user F is 90.
  • the embodiment of the present disclosure first obtains the proportion of the user traffic of all the users that belong to the first IDC, and selects the users that need to be transferred from all the users of the first IDC, wherein the proportion of the user traffic of the user who needs to be transferred The sum is not less than the difference between the peak traffic ratio and the allocated traffic ratio of the first IDC.
  • the first embodiment of the present disclosure randomly selects M users according to the proportion of user traffic of each user, and the sum of user traffic ratios of the M users is not less than the peak traffic proportion and allocation of the first IDC. The difference in traffic ratio.
  • IDC1 is the first IDC
  • the user belonging to IDC1 has user A, user B, user C, User D, user E, user F, and user G
  • user A's user traffic ratio is 5%
  • user B's user traffic is 2%
  • user C's user traffic is 10%
  • user D's user The traffic accounted for 6%
  • the user traffic of user E is 9%
  • the user traffic of user F is 3%. If the difference between the peak traffic of IDC1 and the allocated traffic is 8%, then user A User B, user C, user D, user E, user F, and user G select user A and user F as users who need to transfer according to their corresponding user traffic ratio.
  • Manner 2 According to the proportion of the user traffic, the top N users ranked from high to low are used as the user who needs to be transferred; where N is a positive integer.
  • the sum of the user traffic ratios of the selected N users is not less than the difference between the peak traffic proportion of the first IDC and the proportion of the allocated traffic.
  • the user who needs to transfer when the user who needs to transfer is determined by using the second method, all the users of the first IDC need to be sorted according to the proportion of the user traffic, and when the user who needs to transfer needs to be determined, The N users are selected in a small order, and the sum of the user traffic ratios of the N users is not less than the difference between the peak traffic ratio and the allocated traffic ratio of the first IDC.
  • the IDC1 is the first IDC
  • the users belonging to the IDC1 are the user A, the user B, the user C, the user D, the user E, the user F, and the user G
  • the user traffic of the user A is 5%
  • the user B is the user.
  • the user traffic of user C accounted for 10%
  • the user traffic of user D accounted for 6%
  • the user traffic of user E accounted for 9%
  • the user traffic of user F accounted for 3%.
  • the difference between DC1's peak traffic and distribution traffic is 13%.
  • all users of the home IDC1 are ranked according to the proportion of user traffic, which are user C, user E, user D, user A, user F, and user B, and then from the home IDC1. Two users are selected, namely user C and user E, and user C and user E are regarded as users who need to be transferred.
  • the embodiment of the present disclosure After determining the user in the first IDC that needs to be transferred, the embodiment of the present disclosure also needs to determine the second IDC that can accept the user in the first IDC that needs to be transferred.
  • a first IDC After determining the user in the first IDC that needs to be transferred.
  • the user who needs to transfer in the IDC needs to be assigned a second IDC, that is, it can be determined from all the second IDCs. At least one second IDC of the user in the first IDC that needs to be transferred is included.
  • the set transfer principle from all the second IDCs capable of accepting users, it is determined that at least one second IDC of the user in the first IDC that needs to be transferred is accepted.
  • the transfer principle set by the embodiment of the present disclosure includes one of the following transfer principles.
  • Transfer principle 1 Sort all the first IDCs according to the first difference of all the first IDCs from large to small; and according to the second difference of all the second IDCs, all of the said The second IDC is sorted according to the largest to the smallest;
  • the second IDC that matches the first IDC Determining, for any one of the first IDCs, the second IDC that matches the first IDC, wherein the second IDC that matches the first IDC is in all of the sorted second IDCs a position not lower than a position of the second IDC matching the target first IDC in all of the sorted second IDCs, the target first IDC being low in all of the sorted first IDCs a location in the first IDC of the first IDC after sorting;
  • the first difference of the first IDC is the difference between the peak traffic ratio of the first IDC and the allocated traffic of the first IDC; the second difference of the second IDC is The difference between the allocated traffic ratio of the second IDC and the peak traffic of the second IDC.
  • each first IDC is arranged according to the size of the first difference, from large to small; and, according to each second IDC, each second IDC is sorted according to the second difference size, from large to small; After all the first IDCs and all the second IDCs are sorted, according to the ordering of the first IDC, the users in the first IDC that need to be transferred are sequentially transferred, and the users in the first IDC that need to be transferred are preferentially transferred to the prioritized The second IDC.
  • the first IDC includes IDC1, IDC2, IDC3, and IDC4, and the second IDC includes IDC5, IDC6, and IDC7, and the peak traffic ratio of IDC1 is 15%, and the allocated traffic percentage is 10%; the peak traffic ratio of IDC2 is 21%, the distribution traffic accounted for 15%; IDC3 peak traffic accounted for 28%, The proportion of distributed traffic is 20%; the peak traffic of IDC4 is 13%, the proportion of allocated traffic is 10%; the peak traffic of IDC5 is 8%, the proportion of allocated traffic is 15%; the peak traffic of IDC6 is For 10%, the proportion of allocated traffic is 20%; the peak traffic of IDC7 is 5%, and the proportion of allocated traffic is 10%; then the first difference of IDC1 is 5%, and the first difference of IDC2 is 6.
  • IDC3 has a first difference of 8%
  • IDC4 has a first difference of 3%
  • IDC5 has a second difference of 7%
  • IDC6 has a second difference of 10%
  • IDC7 has a second difference of 5%. %.
  • all the first IDCs are sorted into: IDC3, IDC2, IDC1, IDC4; all the second IDCs are sorted into: IDC6, IDC5, IDC7.
  • the user who needs to transfer 4% of the user traffic is transferred to IDC5; then the user who needs to transfer the user traffic in IDC1 is transferred to IDC5, and the user traffic that needs to be transferred in IDC1 is transferred.
  • the user whose ratio is 2% is transferred to IDC7; finally, the user who needs to transfer the user traffic in IDC4 to 3% is transferred to IDC7, thereby realizing the user's transfer.
  • the second principle is that the second IDC that is closest to the first IDC in the second IDC that can accept the user is used as the second IDC that can receive the user who needs to be transferred in the first IDC.
  • the embodiment of the present disclosure is directed to a first IDC. After determining the user in the first IDC that needs to be transferred, in all the second IDCs capable of accepting the user, determining the IDC closest to the user who needs to be transferred, the first The user in the IDC who needs to transfer is transferred to the determined second IDC.
  • the users are sorted according to the size of the traffic proportion. According to the order of the users that need to be transferred, the users with a large proportion of traffic are preferentially transferred out.
  • the user needs to be transferred to the second IDC closest to the user according to the distance that the user can receive the second IDC of the user.
  • the embodiment of the present disclosure may prioritize users that need to be transferred in the first IDC with a larger difference according to the difference size of all the first IDCs. Transfer.
  • the embodiment of the present disclosure may also determine the transfer order of the first IDC according to other principles. The embodiment of the present disclosure does not limit the transfer order of the first IDC when transferring the user according to the transfer principle 2.
  • the first IDC includes IDC1, IDC2, IDC3, and IDC4, and the second IDC includes IDC5, IDC6, and IDC7, and the peak traffic ratio of IDC1 is 15%, and the allocated traffic percentage is 10%; the peak traffic ratio of IDC2 is 21%, the distribution of traffic accounted for 15%; IDC3 peak traffic accounted for 28%, allocated traffic accounted for 20%; IDC4 peak traffic accounted for 13%, allocated traffic accounted for 10%; IDC5 peak Traffic accounted for 8%, allocated traffic accounted for 15%; IDC6 peak traffic accounted for 10%, allocated traffic accounted for 20%; IDC7 peak traffic accounted for 5%, allocated traffic accounted for 10% Then, it is determined that the first difference of IDC1 is 5%, the first difference of IDC2 is 6%, the first difference of IDC3 is 8%, the first difference of IDC4 is 3%, and the second difference of IDC5 is 7%, the second difference of IDC6 is 10%, and the second difference of IDC7 is 5%.
  • the first IDC may be sorted according to the difference value of the first IDC in descending order, that is, IDC3, IDC2, IDC1, and IDC4. Then, when the user is transferred, the user who needs to transfer in IDC3 is preferentially transferred. For example, when transferring users that need to be transferred in IDC1, it is assumed that the users that need to be transferred in IDC1 are User A, User B, and User C, and the traffic of User A accounts for 4%, and the traffic of User B accounts for 3%. User C's traffic ratio is 1%.
  • IDC1 users When transferring IDC1 users, user A is preferentially transferred, and it is determined that user A's second IDC, IDC5, IDC6, and IDC7, can be accepted, and distances are selected from IDC5, IDC6, and IDC7.
  • User A's recent IDC assuming IDC7, transfers user A to IDC7; when transferring user B, since IDC7 can only accept 1% of users, the second IDC that can accept user B at this time includes IDC5 and IDC6.
  • the user B After determining that the IDC closest to the user B is the IDC5, the user B is transferred to the IDC5; when the user C is transferred, the user C has a traffic ratio of 1%, and the IDC5 can also accept the user with a traffic percentage of 4%.
  • IDC6 can also accept users with a traffic percentage of 10%.
  • IDC7 can also accept users with a traffic percentage of 1%.
  • the second IDC that can accept user C includes IDC5, IDC6, and IDC7, and determines the distance from user C. If the latest IDC is IDC5, user C will be transferred to IDC5. .
  • the embodiment of the present disclosure performs a transfer process on the determined user, wherein the transfer process includes but is not limited to:
  • the user data of the user is migrated from the IDC before the transfer to the IDC of the user after the transfer, and the IDC to which the user belongs in the routing table is modified.
  • the user data of the user is migrated from the IDC before the transfer to the IDC after the user transfer, and part or all of the user data recorded in the IDC before the migration is migrated to the IDC after the user is transferred, and the user is about to transfer.
  • the user data recorded in the previous IDC about the user is deleted.
  • the IDC of the embodiment of the present disclosure includes IDC1, IDC2, IDC3, and IDC4, and the user includes user A, user B, user C, user D, user E, user F, user G, user H, user I, and user J, Among them, A, B, C, D, E, F, G, H, I, and J are user IDs.
  • the mapping between the user and the IDC is as shown in FIG. 6A, and the user A, the user B, and the user E belong to the IDC1; the user C, the user F, and the user G, before the user shifts according to the peak traffic ratio and the allocated traffic ratio of the IDC. Attributable to IDC2; User D and User H belong to IDC3; User I and User J belong to IDC4.
  • the set of the mapping relationship between all users and the IDC to which the user belongs is represented by a routing table, as shown in FIG. 6B, which is: Set ⁇ User A, IDC1>, ⁇ User B, IDC1>, ⁇ User C, IDC2>, ⁇ User D, IDC3>, ⁇ User E, IDC1>, ⁇ User F, IDC2>, ⁇ User G, IDC2>, ⁇ User H, IDC3>, ⁇ User I, IDC4>, ⁇ User J, IDC4 >>.
  • IDC1 and IDC2 are the first IDC
  • IDC3 and IDC4 are the second IDC
  • user A in IDC1 needs to be transferred to IDC3
  • IDC2 is to be User C and user G are transferred to IDC4
  • user A and user F are transferred, the mapping relationship between user and IDC is as shown in FIG. 6C
  • user B and user E are attributed to IDC1
  • user C and user G are attributed to IDC2
  • User A, User D, and User H belong to IDC3
  • User F, User I, and User J belong to IDC4. Modify the IDC of the user in the routing table.
  • the modified routing table is as shown in Figure 6D: Set ⁇ User A, IDC3>, ⁇ User B, IDC1>, ⁇ User C, IDC2>, ⁇ User D, IDC3 >, ⁇ User E, IDC1>, ⁇ User F, IDC4>, ⁇ User G, IDC2>, ⁇ User H, IDC3>, ⁇ User I, IDC4>, ⁇ User J, IDC4>>.
  • an embodiment of the present disclosure provides an overall flow chart of a method for assigning an IDC to a user.
  • Step 701 Determine an allocation ratio of each IDC, and determine a peak traffic proportion of each IDC.
  • Step 702 Determine a first IDC that needs to be transferred by the user, and a second IDC that can receive the user, where the first IDC is greater than the proportion of the allocated traffic, and the second IDC is less than the allocated traffic. ratio;
  • Step 703 Determine, for any one of the first IDCs, all users that belong to the first IDC according to the binding relationship between the IDC and the user.
  • Step 704 Determine a proportion of user traffic of all users that belong to the first IDC.
  • Step 705 According to the proportion of the user traffic, the top N users that are sorted from high to low are used as the users that need to be transferred, and the sum of the user traffic proportions of the selected users is not less than the first IDC. The difference between the peak traffic ratio and the allocated traffic ratio;
  • Step 706 Sort all the first IDCs according to the first difference of all the first IDCs from large to small; and according to the second difference of all the second IDCs, perform all the second IDCs from large to small. Sorting; for any one of the first IDCs, determining a second IDC that matches the first IDC; wherein the second IDC that matches the first IDC is not lower than the target first in all of the sorted second IDCs The position of the IDC matching second IDC in all the second IDCs after sorting, the position of the target first IDC in all the first IDCs after sorting is lower than all the contents of the first IDC after sorting Describe the location in the first IDC;
  • Step 707 will be able to receive the second IDC of the second IDC of the user that is closest to the first IDC, as the second IDC capable of accepting the user in the first IDC that needs to be transferred;
  • Step 708 Perform a transfer process on the determined user.
  • the time period in which the embodiment of the present disclosure allocates the IDC to the user includes a busy period and a non-busy period.
  • the busy period is a peak period of expected traffic;
  • the non-busy period is a period other than the busy period in daily situations.
  • the busy period is a peak period of expected traffic;
  • the non-busy period is a period other than the busy period in daily situations.
  • the busy period is a peak period of expected traffic;
  • the non-busy period is a period other than the busy period in daily situations.
  • the busy period is a peak period of expected traffic;
  • the non-busy period is a period other than the busy period in daily situations. For example, Facebook Group’s global double eleven shopping carnival, On November 11th, the number of users visiting the Facebook increased suddenly, and the traffic weight generated by each user suddenly increased, and the period was a busy period.
  • the embodiment of the present disclosure may adopt the foregoing method for allocating an IDC to a user during a busy period, and determine the first IDC and the second IDC according to the peak traffic proportion and the allocated traffic ratio of each IDC, and implement the first Some users of an IDC are transferred to the second IDC.
  • the IDC is reasonably allocated to the user to fully meet the needs of users during busy periods.
  • Embodiments of the present disclosure allocate IDCs to users in a non-busy time period.
  • the traffic weights generated by the users belonging to the IDC are small. Under normal circumstances, the actual traffic generated by each IDC is much smaller than the allocated traffic of the IDC.
  • Each IDC can meet the needs of users belonging to the IDC. In order to improve the user experience, the IDC is assigned to the user.
  • the embodiment of the present disclosure determines the attribution of the user's continuous login according to the ID of the user, determines the IDC corresponding to the user according to the attribution of the user's continuous login, and associates the IDC of the user with the user. It is represented by the mapping relationship between the user ID and the IDC.
  • the ID of the user of the embodiment of the present disclosure may be a unique identity ID generated for the user when the user registers on the Internet.
  • a method for generating a binding relationship between an IDC and a user includes:
  • Step 801 Determine, according to an access log of the user at different times, a home location when the user logs in.
  • Step 802 Determine the number of occurrences of the same home location from the home location when the user logs in multiple times.
  • Step 803 Determine an IDC bound by the user according to the most-origined attribution, and generate a binding relationship between the data center and the user.
  • the embodiment of the present disclosure When determining the IDC to which the user belongs, the embodiment of the present disclosure first determines the attribution of the user. Embodiments of the present disclosure determine the attribution of a user using the following method.
  • the embodiment of the present disclosure obtains the historical access log of the user, uses DNS technology, parses the attribution of the user's historical login, and determines the attribution of the user through big data analysis.
  • all the access logs of the user for one month can be obtained, and the number of times each home location is parsed is counted based on the home location corresponding to all access logs parsed by the DNS, and the attribution site with the most occurrences is used as the attribution of the user. Ground.
  • the access log of user A for one month is obtained, and the access log of user A for one month is subjected to DNS resolution, and the attribution domain corresponding to the access log of user A for one month is determined, and each home location is parsed. The number of times. If the user A is located in Beijing and Hangzhou at the time of login within one month, and according to the access log, it is determined that User A logs in 25 times in Beijing and Hangzhou logs in 3 times, and it is determined that User A's attribution is Beijing.
  • the embodiment of the present disclosure determines the IDC of the user binding according to the attribution of the user after determining the attribution of the user. Specifically, the embodiment of the present disclosure takes the IDC closest to the home of the user as the IDC bound by the user.
  • the access log of user A for one month is obtained, and the access log of user A for one month is subjected to DNS resolution, and the attribution domain corresponding to the access log of user A for one month is determined, and each home location is parsed. The number of times. If the user A is located in Beijing and Hangzhou at the time of login within one month, and according to the access log, it is determined that User A logs in 25 times in Beijing and Hangzhou logs in 3 times, and it is determined that User A's attribution is Beijing. And the IDC closest to Beijing is IDC2, and it is determined that the IDC corresponding to User A is IDC2.
  • the mapping relationship between each user and its belonging IDC is taken as a set, and the set is a routing table.
  • the IDC corresponding to user A is IDC1
  • the IDC corresponding to user B is IDC1
  • the IDC corresponding to user C is IDC2
  • the IDC corresponding to user D is IDC3
  • the IDC corresponding to user E is IDC3
  • the IDC corresponding to user F is IDC4.
  • a routing table in the form of a routing table is expressed as: Set ⁇ user A, IDC1>, ⁇ user B, IDC1>, ⁇ user C, IDC2>, ⁇ user D, IDC3>, ⁇ user E, IDC3>, ⁇ use User F, IDC4>>.
  • the embodiment of the present disclosure determines the home location when the user logs in according to the IP address of the user, and determines the IDC corresponding to the home location when the user logs in, and compares the attribution corresponding to the home location when the user logs in.
  • the IDC and the IDC currently bound by the user determine how the IDC provides services to the user based on the comparison result.
  • the IDC corresponding to the home location when the user logs in is the same as the IDC bound by the user, and the network service is provided to the user by the IDC corresponding to the home location when the user logs in.
  • the IDC corresponding to Hangzhou is IDC3.
  • the IDC currently bound by user B is IDC3.
  • the IDC corresponding to the attribution of the user B when the user B logs in is the same IDC as the IDC currently bound by the user, and the network service is directly provided to the user through the IDC3.
  • the IDC corresponding to the home location when the user logs in is different from the IDC currently bound by the user, and the user is served by the IDC currently bound by the user.
  • the current bound IDC is controlled to provide the network service for the user.
  • the IDC of Hangzhou is determined to be IDC3.
  • the IDC currently bound by user B is IDC4. Then, it is determined that the IDC corresponding to the attribution of the user B when the user B is logged in is not the same IDC as the IDC currently bound by the user, and the network service is provided to the user through the IDC4 currently bound by the user B.
  • the embodiment of the present disclosure further provides a method for an IDC to provide a service for a user, including:
  • Step 901 Obtain an access log of the user when determining that the user logs in.
  • Step 902 Using DNS technology, determining a home location when the user logs in according to the user IP address;
  • Step 903 Call a routing table, and determine an IDC currently bound by the user according to the routing table.
  • Step 904 it is determined whether the home location of the user login is the same as the home location of the IDC currently bound by the user; if yes, step 905 is performed; if not, step 906 is performed;
  • Step 905 Control an IDC corresponding to the home location when the user logs in to provide a network service for the user.
  • Step 906 Jump to the IDC currently bound by the user, and control the IDC currently bound by the user to provide network services for the user.
  • the user When the user logs in, the user first determines the IDC corresponding to the home location when the user logs in according to the access log of the user. If the IDC corresponding to the home location when the user logs in is determined, the IDC currently bound to the user is determined. At the same time, it jumps to the IDC currently bound by the user, and controls the IDC currently bound by the user to provide network services for the user. However, if it is determined that the IDC corresponding to the home location of the user is the same every time the user logs in, and the IDC currently bound by the user is different, the IDC currently bound by the user needs to be modified.
  • the IDC corresponding to the home location when the user logs in is determined to be the same within the set duration according to the access log of the user at different times, and is different from the IDC currently bound by the user, The IDC corresponding to the home location when the user logs in, and the IDC currently bound by the user is modified.
  • the user each time the user logs in, the user first determines the attribution of the user according to the IP address of the user according to the access log of the user, and determines the IDC corresponding to the attribution of the user.
  • the IDC is the IDC closest to the home of the user's login.
  • the routing table indicating the binding relationship between the IDC and the user needs to be invoked, and the routing table is used to determine whether the IDC closest to the user is the same as the IDC currently bound by the user in the routing table.
  • the IDC closest to the user is not the same IDC as the IDC currently bound by the user in the routing table, and within the set duration, the latest IDC is the same for each login, and the IDC currently bound to the user. If the IDC is not the same ID, the user's attribution is changed. In this case, the IDC currently bound by the user needs to be modified.
  • the condition for determining the IDC currently bound by the user is:
  • the IDC corresponding to the home location when the user logs in is determined to be the same one, and the home location corresponding to the user login is corresponding to the access log of the user at different times.
  • the IDC is different from the IDC currently bound by the user, and the user is logged in.
  • the data center corresponding to the attribution is used as a data center bound by the user.
  • the set duration of the embodiments of the present disclosure may be 30 days, or 50 days.
  • the embodiment of the present disclosure does not limit the set duration.
  • user A is an Facebook Chinese user, and the IDC currently bound by user A is IDC1, where IDC1 is a server located in China; after user A immigrates to the US, when user A logs in to the Facebook website in the United States, according to user A
  • the access log determines that the IDC closest to the user A is the IDC5 of the US according to the IP address of the user A.
  • the jump to the IDC1 provides the network for the user A. service.
  • IDC5 the IDC5 of the US according to the IP address of the user A.
  • the user is transferred to the changed data center.
  • the transfer processing of the user in the embodiment of the present disclosure includes migrating part or all of the user data of the user, that is, migrating part or all of the user data of the user from the originally bound IDC to the modified IDC.
  • the embodiment of the present disclosure may first modify the home IDC corresponding to the user in the routing table, and then migrate part or all of the user data to the user.
  • the modified IDC may also be: first, part or all of the user data of the user is migrated to the modified IDC, and then the home IDC corresponding to the user in the routing table is modified.
  • the embodiment of the present disclosure does not limit the order in which the IDC currently bound by the user and the user are transferred.
  • user A is an Facebook Chinese user, and the IDC currently bound by user A is IDC1, where IDC1 is a server located in China; after user A immigrates to the US, when user A logs in to the Facebook website in the United States, according to user A
  • the access log determines that the IDC closest to the user A is the IDC5 of the US according to the IP address of the user A.
  • the jump to the IDC1 provides the network for the user A. service.
  • the IDC of each user's initial attribution may be determined by the method for allocating the IDC to the user during the non-busy period provided by the embodiment of the present disclosure, or may be determined by other methods.
  • an embodiment of the present disclosure provides a device for allocating a data center to a user. Since the principle of solving the problem is similar to the method for allocating a data center to a user in the embodiment of the present disclosure, the device is Implementation can refer to the implementation of the method, and the repetition will not be repeated.
  • an embodiment of the present disclosure provides a device for allocating a data center to a user, including:
  • the obtaining module 1001 is configured to determine an allocated traffic proportion of each data center, and determine a peak traffic proportion of each of the data centers;
  • a determining module 1002 configured to determine a first data center that needs to perform user transfer and a second data center that can receive the user;
  • the processing module 1003 is configured to process, in the first data center, a user that needs to be transferred to the determined second data center, where the first data center is greater than the allocated traffic.
  • the second data center is a data center in which the peak traffic ratio is smaller than the proportion of the allocated traffic.
  • the ratio of the allocated traffic of the data center is a ratio of an upper limit of the data center carrying user traffic to an upper limit of the data center carrying user traffic.
  • the obtaining module 1001 is specifically configured to:
  • the ratio of the sum of the traffic weights of the users belonging to the data center to the sum of the traffic weights of all users belonging to all data centers is used as the peak traffic of the data center. Proportion.
  • the traffic weight of the user is a value indicating a proportion of user traffic generated by the user to a sum of user traffic generated by all users of all data centers.
  • processing module 1003 is specifically configured to:
  • processing module 1003 is specifically configured to:
  • Determining a proportion of user traffic belonging to each user of the first data center selecting, according to the user traffic ratio, all users that need to be transferred from all the users belonging to the first data center, wherein the selected ones are selected
  • the sum of the user traffic proportions of the user is not less than a difference between the peak traffic proportion of the first data center and the allocated traffic ratio.
  • processing module 1003 is specifically configured to:
  • the top N users ranked from high to low are used as the users who need to transfer; wherein N is a positive integer.
  • processing module 1003 is specifically configured to:
  • the transfer principle includes one of the following transfer principles:
  • Transfer principle 1 Sort all the first data centers according to the first difference of all the first data centers, from large to small; and according to the second difference of all the second data centers, All of the second data centers are sorted from large to small; for any one of the first data centers, the second data center that matches the first data center is determined, wherein the first data center The matched second data center is located in all of the sorted second data centers at a position not lower than all of the sorted contents of the second data center matching the target first data center a location in the data center, the location of the target first data center in all of the sorted first data centers is lower than the location of the first data center in all of the sorted first data centers
  • the first difference of the first data center is a difference between a peak traffic ratio of the first data center and an allocated traffic ratio of the first data center; The difference is the difference between the ratio of the allocated traffic of the second data center and the peak traffic of the second data center;
  • processing module 1003 is further configured to generate, according to the following manner, a binding relationship between the data center and a user:
  • the step of determining the first data center that needs to perform user transfer and the second data center capable of receiving the user is completed during a busy time period;
  • the step of determining the attribution of the user multiple times according to the access log of the user at different times is completed in a non-busy time period.
  • processing module 1003 is further configured to:
  • the data center corresponding to the home location when the user logs in is determined to be the same one, and the user logs in.
  • the data center corresponding to the home is different from the data center to which the user is currently bound, and the data center corresponding to the home address when the user logs in is used as the data center bound by the user.
  • processing module 1003 is further configured to:
  • processing module 1003 is further configured to:
  • the currently bound data center is controlled to provide network services for the user.
  • the present disclosure may also be implemented in hardware and/or software (including firmware, resident software, microcode, etc.). Still further, the present disclosure may take the form of a computer program product on a computer usable or computer readable storage medium having computer usable or computer readable program code embodied in a medium for use by an instruction execution system Or used in conjunction with the instruction execution system.
  • a computer usable or computer readable medium can be any medium that can contain, store, communicate, communicate, or transport a program for use by an instruction execution system, apparatus, or device, or in conjunction with an instruction execution system. , device or device use.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本公开内容涉及网络通信技术领域，特别涉及一种为用户分配数据中心的方法和设备，用以解决现有技术基于IP地址段为用户分配数据中心的方法，在流量高峰期时，不能合理的为用户分配数据中心的问题。本公开内容实施例确定每个数据中心的分配流量占比，以及确定所述每个数据中心的高峰流量占比；确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心；将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心；由于本公开内容实施例在流量高峰期时，调整用户归属的数据中心，能够为用户合理的分配数据中心。

Description

一种为用户分配数据中心的方法和设备 技术领域
本公开内容涉及网络通信技术领域，特别涉及一种为用户分配数据中心的方法和设备。
背景技术
随着互联网技术的发展，互联网的用户越来越多，分布也越来越广泛。尤其大互联网公司，其用户大都分布在广泛的地理区域，而且越来越多的大型互联网公司将业务向全球发展，这使得其提供的网络服务面临着分布全球的海量用户。
随着互联网公司用户的增多，以及用户分布越来越广泛，互联网为了更好地满足用户需要，需要提供多个数据中心。由于用户的访问速度往往因归属数据中心的地点而不同，理论上用户归属的数据中心距离用户越近，访问速度越快。因此，在为用户分配归属的数据中心时，需要将用户分配至距离其最近的数据中心。互联网公司往往会面临可预期的突发流量增长，在面临突发流量增长时，由于不同数据中心能够承载用户产生的流量是不同的，有的数据中心无法承载流量高峰期就近访问的用户产生的流量，这就需要在突发流量增长时，合理分配归属于数据中心的用户。比如阿里巴巴集团每年一次的全球双十一购物狂欢节。如果采用就近分配数据中心的方法为用户分配数据中心，若大多数狂欢节的用户都集中在某一个区域，则会出现局部流量过大的问题。
现有技术在确定用户归属的数据中心时，基于DNS(Domain Name System，域名系统)技术，根据用户IP(Internet Protocol，网络之间互联的协议)地址段所归属的地理区域，将域名解析到最近的机房，从而为用户就近分配数据中心。但是，由于IP地址与用户的对应关系非常间接，多个用户可能共用同一个IP地址，一个用户也可能使用多个IP地址。基于IP地址段为用户分配数据 中心的方法，在流量高峰期时，无法准确确定每个数据中心的实际用户数量，不能合理的为用户分配数据中心。
综上所述，目前的基于IP地址段为用户分配数据中心的方法，在流量高峰期时，不能合理的为用户分配数据中心。
发明内容
本公开内容提供一种为用户分配数据中心的方法和设备，用以解决现有技术基于IP地址段为用户分配数据中心的方法，在流量高峰期时，不能合理的为用户分配数据中心的问题。
基于上述问题，本公开内容实施例提供一种为用户分配数据中心的方法，包括：
确定每个数据中心的分配流量占比，以及确定所述每个数据中心的高峰流量占比；
确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心；
将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心；
其中所述第一数据中心为所述高峰流量占比大于所述分配流量占比的数据中心，所述第二数据中心为所述高峰流量占比小于所述分配流量占比的数据中心。
由于本公开内容实施例将高峰流量占比大于分配流量占比的数据中心作为第一数据中心，并确定第一数据中心中需要转移的用户，以及将分配流量占比大于高峰流量占比的数据中心作为第二数据中心，根据所述第二数据中心，对所述第一数据中心中需要转移的用户进行转移处理，确保每个数据中心都能满足归属于该数据中心的所有用户使用。本公开内容实施例根据每个数据中心的分配流量占比和高峰流量占比，在数据中心的分配流量占比和高峰流量占比存在差值时，调整用户归属的数据中心，为用户合理的分配数据中心。
可选的，所述数据中心的分配流量占比为所述数据中心承载用户流量的上限占全部数据中心承载用户流量的上限之和的比值。
可选的，所述确定所述每个数据中心的高峰流量占比，包括：
针对任意一个所述数据中心，将归属所述数据中心的用户的流量权值之和，与归属全部数据中心的所有用户的所述流量权值之和的比值，作为所述数据中心的高峰流量占比。
由于本公开内容实施例提供了根据流量权值确定数据中心的高峰流量占比的方法，从而准确计算得到数据中心的高峰流量占比。
可选的，所述用户的流量权值为表示所述用户产生的用户流量占全部数据中心的所有用户产生的用户流量之和的比重的数值。
由于本公开内容实施例给出了预测用户的流量权值的方法，从而能够根据用户的流量权值得到数据中心的高峰流量占比。
可选的，所述将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心，包括：
针对任意一个所述第一数据中心，根据数据中心和用户的绑定关系，确定归属所述第一数据中心的所有用户；
从归属所述第一数据中心的所述所有用户中，确定所述第一数据中心中需要进行转移的用户，以及从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个所述第二数据中心；并将所述需要进行转移的用户转移处理到所述确定的至少一个所述第二数据中心。
由于本公开内容实施例针对一个第一数据中心，从该第一数据中心中选取需要进行转移的用户，将第一数据中心中需要转移的用户转移至第二数据中心中，确保每个数据中心都能满足归属于该数据中心的所有用户使用。
可选的，所述确定所述第一数据中心中需要进行转移的用户，包括：
确定归属所述第一数据中心的每一个用户的用户流量占比；
根据所述用户流量占比，从归属所述第一数据中心的所有所述用户中选取需要进行转移的用户，其中选取的所述用户的所述用户流量占比之和不小于所述第一数据中心的所述高峰流量占比与所述分配流量占比之差。
由于本公开内容实施例根据第一数据中心的每一个用户的流量占比，在确定需要将第一数据中心中需要转移的用户进行转移时，从第一数据中心中选取出流量占比之和不小于第一数据中心的高峰流量占比与分配流量占比之差的用户，使得第一数据中心能够满足归属于该第一数据中心的用户使用，从而合理的为用户分配数据中心。
可选的，所述根据所述用户流量占比，从归属所述第一数据中心的所有所述用户中选取需要进行转移的用户，包括：
根据所述用户流量占比，将从高到低排序的前N个用户作为所述需要进行转移的用户；
其中N为正整数。
由于本公开内容实施例在确定第一数据中心中需要进行转移的用户时，根据第一数据中心中所有用户的用户流量占比，选取出用户流量占比最大的N个用户作为需要进行转移的用户，并且该选取出的N个用户的用户流量占比之和不小于该第一数据中心的高峰流量占比与分配流量占比之差，从而在影响尽量少的用户的情况下，实现数据中心的合理分配。
可选的，从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个所述第二数据中心，包括：
根据设定的转移原则，从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个所述第二数据中心；
所述转移原则包括下列转移原则中的一个：
转移原则一、根据所有所述第一数据中心的第一差值，将所有所述第一数 据中心按照从大到小进行排序；以及根据所有所述第二数据中心的第二差值，将所有所述第二数据中心按照从大到小进行排序；
针对任意一个所述第一数据中心，确定与所述第一数据中心匹配的所述第二数据中心，其中与所述第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置不低于与目标第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置，所述目标第一数据中心在排序后的所有所述第一数据中心中的位置低于所述第一数据中心在排序后的所有所述第一数据中心中的位置；
其中，所述第一数据中心的第一差值为所述第一数据中心的高峰流量占比和所述第一数据中心的分配流量占比的差值；所述第二数据中心的第二差值为所述第二数据中心的分配流量占比和所述第二数据中心的高峰流量占比的差值；
转移原则二、将所述能够接纳用户的第二数据中心中距离所述第一数据中心最近的第二数据中心，作为能够接纳所述第一数据中心中所述需要进行转移的用户的第二数据中心。
由于本公开内容实施例在确定第一数据中心中需要转移的用户之后，提供了多种转移规则，用于确定接纳第一数据中心需要转移的用户的第二数据中心，以及不同的转移方法，从而实现了为用户灵活的分配数据中心。
可选的，根据下列方式生成所述数据中心和用户的绑定关系：
根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地；
从所述多次确定的所述用户登录时的归属地中确定同一个归属地出现的次数；
根据次数最多的归属地，确定所述用户绑定的数据中心，并生成所述数据中心和用户的绑定关系。
由于本公开内容实施例根据用户ID唯一标识用户，并确定用户长期登录的归属地，将距离用户长期登录的归属地最近的数据中心作为用户归属的数据 中心，避免了根据IP地址确定用户归属的数据中心的不准确性，从而准确地将用户归属到最近的数据中心。
可选的，所述确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心的步骤是在繁忙时段内完成的；
所述根据不同时间的用户的访问日志，多次确定所述用户的归属地的步骤是在非繁忙时段内完成的。
由于本公开内容实施例提供了针对繁忙时期和非繁忙时期，分别为用户分配数据中心的方法。在非繁忙时期，为用户就近分配数据中心；在非繁忙时期，根据每个数据中心的分配流量占比和高峰流量占比，合理调整归属于数据中心的用户，从而灵活准确地为用户分配数据中心。
可选的，所述根据次数最多的归属地，确定所述用户绑定的数据中心之后，还包括：
在非繁忙时段的设定时长内，若根据不同时间的所述用户的访问日志，多次确定所述用户登录时的归属地对应的数据中心为同一个，且所述用户登录时的所述归属地对应的所述数据中心与所述用户当前绑定的数据中心不同，则将所述用户登录时的所述归属地对应的所述数据中心作为所述用户绑定的数据中心。
由于本公开内容实施例在确定用户的归属地发生变化后，多次确定变化后的归属地，若在设定时长内变化后的归属地为同一个，则确定用户的归属地发生变化，修改该用户当前绑定的数据中心，从而更好地满足用户通过最近的数据中心访问互联网。
可选的，该方法还包括：
在非繁忙时段内，若所述数据中心和用户的绑定关系发生变化后，将所述用户转移处理到所述变化后的数据中心。
由于本公开内容实施例在数据中心和用户的绑定关系发生变化后，还需要将用户归属的数据中心发生变化的用户数据进行转移，将用户的用户数据从原 来归属的数据中心迁移至变化后的数据中心中。
可选的，该方法还包括：
在非繁忙时段内，若所述用户登录时的归属地与所述用户当前绑定的数据中心对应的归属地不同，控制所述当前绑定的数据中心为所述用户提供网络服务。
由于本公开内容实施例每个用户的用户数据存放在用户归属的数据中心中，在用户登录地发生变化后，仍然使用用户归属的数据中心为用户提供网络服务，从而避免了同一用户对应多个数据中心的问题，提高了数据中心提供网络服务的准确性。
另一方面，本公开内容实施例还提供一种为用户分配数据中心的设备，包括：
获取模块，用于确定每个数据中心的分配流量占比，以及确定所述每个数据中心的高峰流量占比；
确定模块，用于确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心；
处理模块，用于将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心；其中所述第一数据中心为所述高峰流量占比大于所述分配流量占比的数据中心，所述第二数据中心为所述高峰流量占比小于所述分配流量占比的数据中心。
可选的，所述数据中心的分配流量占比为所述数据中心承载用户流量的上限与全部数据中心承载用户流量的上限之和的比值。
可选的，所述获取模块，具体用于：
针对任意一个所述数据中心，将归属所述数据中心的用户的流量权值之和，与归属全部数据中心的所有用户的所述流量权值之和的比值，作为所述数据中心的高峰流量占比。
可选的，所述处理模块，具体用于：
针对任意一个所述第一数据中心，根据数据中心和用户的绑定关系，确定归属所述第一数据中心的所有用户；从归属所述第一数据中心的所述所有用户中，确定所述第一数据中心中需要进行转移的用户，以及从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个所述第二数据中心；并将所述需要进行转移的用户转移处理到所述确定的至少一个所述第二数据中心。
可选的，所述处理模块，具体用于：
确定归属所述第一数据中心的每一个用户的用户流量占比；根据所述用户流量占比，从归属所述第一数据中心的所有所述用户中选取需要进行转移的用户，其中选取的所述用户的所述用户流量占比之和不小于所述第一数据中心的所述高峰流量占比与所述分配流量占比之差。
可选的，所述处理模块，具体用于：
根据所述用户流量占比，将从高到低排序的前N个用户作为所述需要进行转移的用户；其中N为正整数。
可选的，所述处理模块，具体用于：
根据设定的转移原则，从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个第二数据中心；
所述转移原则包括下列转移原则中的一个：
转移原则一、根据所有所述第一数据中心的第一差值，将所有所述第一数据中心按照从大到小进行排序；以及根据所有所述第二数据中心的第二差值，将所有所述第二数据中心按照从大到小进行排序；
针对任意一个所述第一数据中心，确定与所述第一数据中心匹配的所述第二数据中心，其中与所述第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置不低于与目标第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置，所述目标第一数据中心在 排序后的所有所述第一数据中心中的位置低于所述第一数据中心在排序后的所有所述第一数据中心中的位置；
其中，所述第一数据中心的第一差值为所述第一数据中心的高峰流量占比和所述第一数据中心的分配流量占比的差值；所述第二数据中心的第二差值为所述第二数据中心的分配流量占比和所述第二数据中心的高峰流量占比的差值；
转移原则二、将所述能够接纳用户的第二数据中心中距离所述第一数据中心最近的第二数据中心，作为能够接纳所述第一数据中心中所述需要进行转移的用户的第二数据中心。
可选的，所述处理模块，还用于，根据下列方式生成所述数据中心和用户的绑定关系：
根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地；从所述多次确定的所述用户登录时的归属地中确定同一个归属地出现的次数；根据次数最多的归属地，确定所述用户绑定的数据中心，并生成所述数据中心和用户的绑定关系。
可选的，所述处理模块，具体用于：所述确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心的步骤是在繁忙时段内完成的；
所述根据不同时间的用户的访问日志，多次确定所述用户的归属地的步骤是在非繁忙时段内完成的。
可选的，所述处理模块，还用于：
在非繁忙时段的设定时长内，若根据不同时间的所述用户的访问日志，多次确定所述用户登录时的归属地对应的数据中心为同一个，且所述用户登录时的所述归属地对应的所述数据中心与所述用户当前绑定的数据中心不同，则将所述用户登录时的所述归属地对应的所述数据中心作为所述用户绑定的数据中心。
可选的，所述处理模块，还用于：
在非繁忙时段内，若所述数据中心和用户的绑定关系发生变化后，将所述用户转移处理到所述变化后的数据中心。
可选的，所述处理模块，还用于：
在非繁忙时段内，若所述用户登录时的归属地与所述用户当前绑定的数据中心对应的归属地不同，控制所述当前绑定的数据中心为所述用户提供网络服务。
附图说明
图1A为本公开内容实施例用户与IDC的映射关系示意图一；
图1B为本公开内容实施例路由表一的示意图；
图2为本公开内容实施例提供的一种为用户分配IDC的方法流程图；
图3为本公开内容实施例用户与IDC的映射关系示意图二；
图4为本公开内容实施例用户与IDC的映射关系示意图三；
图5为本公开内容实施例路由表二的示意图；
图6A为本公开内容实施例用户转移前的用户与IDC的映射关系示意图；
图6B为本公开内容实施例用户转移前路由表的示意图；
图6C为本公开内容实施例用户转移后的用户与IDC的映射关系示意图；
图6D为本公开内容实施例用户转移后路由表的示意图；
图7为本公开内容实施例提供的一种为用户分配IDC的方法整体流程图；
图8为本公开内容实施例提供的一种的生成IDC和归属IDC的用户的绑定关系的方法流程图；
图9为本公开内容实施例提供一种的IDC为用户提供服务的方法流程图；
图10为本公开内容实施例提供的一种为用户分配数据中心的设备的结构示意图。
具体实施方式
本公开内容实施例确定每个数据中心的分配流量占比，以及确定所述每个数据中心的高峰流量占比；确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心；将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心；其中所述第一数据中心为所述高峰流量占比大于所述分配流量占比的数据中心，第二数据中心为所述高峰流量占比小于所述分配流量占比的数据中心。
由于本公开内容实施例将高峰流量占比大于分配流量占比的数据中心作为第一数据中心，并确定第一数据中心中需要转移的用户，以及将分配流量占比大于高峰流量占比的数据中心作为第二数据中心；根据所述第二数据中心，对所述第一数据中心中需要转移的用户进行转移处理，确保每个数据中心都能满足归属于该数据中心的所有用户使用。本公开内容实施例根据每个数据中心的分配流量占比和高峰流量占比，在数据中心的分配流量占比和高峰流量占比存在差值时，调整用户归属的数据中心，为用户合理的分配数据中心。
本公开内容实施例可以应用在有多个数据中心为用户提供服务的场景。在该场景中，需要根据多个数据中心，为用户合理分配数据中心，以提高用户体验。
本公开内容实施例的数据中心可以为IDC(Internet Data Center，互联网数据中心)，还可以是其它用于存放数据的中心。为了方便介绍，下面以数据中心是IDC为例进行说明。需要说明的是，数据中心是其他用于存放数据的中心的处理方法与数据中心是IDC的处理方法相同，在此不再赘述。
本公开内容实施例的IDC具体为互联网数据中心，在该互联网数据中心中包括了一定数量的网络服务器，并在网络服务器上部署特定的互联网软件，对互联网的用户提供网络服务。
本公开内容实施例包括一个重要的组成部分就是记录用户到其最近IDC的映射关系，又可以称为路由表。具体的，用户到其最近IDC的映射关系以集 合的形式实现，该集合中元素为用户与用户归属的IDC的映射关系，形如<user_id，IDC>。所有的用户与用户归属的IDC的映射关系构成的集合为路由表，路由表可以采用Set<<user_id，IDC>>表示。
例如，本公开内容实施例的IDC包括IDC1、IDC2和IDC3，用户包括用户A、用户B、用户C、用户D，用户E、用户F、用户G和用户H。其中，A、B、C、D、E、F、G和H为用户ID，用户和IDC的映射关系如图1A所示，用户A、用户B和用户E归属于IDC1；用户C、用户F和用户G归属于IDC2；用户D和用户H归属于IDC3。则所有用户与用户归属的IDC的映射关系构成的集合，采用路由表的形式表示，路由表如图1B所示，即：Set<<用户A，IDC1>，<用户B，IDC1>，<用户C，IDC2>，<用户D，IDC3>，<用户E，IDC1>，<用户F，IDC2>，<用户G，IDC2>，<用户H，IDC3>>。
下面结合说明书附图对本公开内容实施例作进一步详细描述。
如图2所示，本公开内容实施例提供的一种为用户分配IDC的方法，包括：
步骤201、确定每个IDC的分配流量占比，以及确定所述每个IDC的高峰流量占比；
步骤202、确定需要进行用户转移的第一IDC和能够接纳用户的第二IDC；
步骤203、将所述第一IDC中需要转移的用户转移处理到所述确定的第二IDC；其中所述第一IDC为所述高峰流量占比大于所述分配流量占比的IDC，所述第二IDC为所述高峰流量占比小于所述分配流量占比的IDC。
本公开内容实施例对于互联网上的每一个用户，均为其分配了唯一标识ID，本公开内容实施例根据用户ID标识用户。例如，对于阿里巴巴互联网的用户，在用户使用阿里巴巴互联网时，会要求用户在其网站上进行注册，在用户注册成功时，为用户分配用户ID。
本公开内容实施例的数据中心的分配流量占比为所述数据中心承载用户流量的上限与全部数据中心承载用户流量的上限之和的比值。
并且，本公开内容实施例数据中心的分配流量占比为预先配置的。实施中， 针对每一个IDC，为每一个IDC预先分配了分配流量占比。比如，<IDC1，30％>，则表示在IDC1中预先归属了30％的流量，即IDC1最多承载全局30％流量。
本公开内容实施例在需要确定IDC的分配流量占比时，可以直接获取得到为每一个IDC预先分配的分配流量占比。
本公开内容实施例预先配置分配流量占比时，根据每个IDC对应的网络服务器的数量确定的。具体的，针对任意一个IDC，在确定该IDC的分配流量占比时，将该IDC对应的网络服务器的数量与全部IDC对应的所有网络服务器的数量的比值作为该IDC的分配流量占比。
例如，本公开内容实施例共包括三个IDC，分别为IDC1、IDC2和IDC3，IDC1对应的网络服务器有30个，IDC2对应的网络服务器有30个，IDC3对应的网络服务器有40个，则确定IDC1和IDC2的分配流量占比为30％，IDC3对应的分配流量占比为40％。
本公开内容实施例的数据中心的高峰流量占比为归属所述数据中心的用户的流量权值之和，与归属全部数据中心的所有用户的所述流量权值之和的比值。
需要说明的是，在确定本公开内容是实施例数据中心的高峰流量占比时，需要预测出归属全部数据中心的每一个用户可能产生的用户流量(在本公开内容实施例中采用用户的流量权值表示)，将该数据中心对应的用户可能产生的用户流量之和，与归属全部数据中心的所有用户可能产生的用户流量之和的比值，作为该数据中心的高峰流量占比。
具体的，本公开内容实施例的每个IDC的高峰流量占比根据下列步骤计算得到。
1、获取所有用户对应的历史访问数据。
本公开内容实施例的所有用户是指使用该互联网的用户，具体的可以是在该互联网上注册过的用户，并且注册过的用户在该互联网上有用户唯一标识ID。例如，阿里巴巴的用户，即在阿里巴巴网络上注册过的会员。
2、通过机器学习算法，确定每一个用户的流量权值。
本公开内容实施例根据获取到的用户对应的历史访问数据，通过机器学习算法，预测出用户可能产生的流量权值。
其中，本公开内容实施例的所述用户的流量权值为表示所述用户产生的用户流量占全部数据中心的所有用户产生的用户流量之和的比重的数值。
其中，用户的流量权值为0-100分的分值，用以标识用户能够产生的流量权重。
本公开内容实施例在预测用户的流量权值时，首先确定用户可能产生的用户流量占全局用户产生的用户流量之和的比值，再根据流量比值与流量权值的对应关系，确定用户的流量权值。
例如，在确定用户A的流量权值时，通过计算得到用户A可能产生的用户流量占全局用户产生的用户流量的比值为0.02，假设流量比值在0.015-0.02之间时，对应的流量权值为80，则确定用户A的流量取值为80。
需要说明的是，本公开内容实施例对于确定用户的流量权值采用的机器算法的具体算法类型不作限定，任何能够根据用户的历史行为数据预测用户可能产生的流量权重的方法均适用于本公开内容。
3、根据每一个用户的流量权值，确定每一个IDC的高峰流量占比。
针对任意一个IDC，将归属所述IDC的用户的流量权值之和，与归属全部IDC的所有用户的流量权值之和的比值，作为所述IDC的高峰流量占比。
具体的，针对任意一个IDC，确定归属于该IDC的所有的用户，将归属于该IDC的所有用户的流量权值相加，得到该IDC的流量权值；将全局所有用户的流量权值相加，得到全局的流量权值，将IDC的流量权值与全局的流量权值的比值，作为IDC的高峰流量占比。
本公开内容实施例每一个IDC的高峰流量占比为每一个IDC可能产生的最大流量占比，在实际过程中，每一个IDC产生的流量占比小于或等于该高峰流量占比。
例如，假设全局共包括三个IDC，分别为IDC1、IDC2和IDC3；全局用户包括用户A、用户B、用户C、用户D、用户E、用户F和用户G，其中，用户和IDC的映射关系如图3所示，归属于IDC1的用户有用户A和用户B；归属于IDC2的用户有用户C、用户D和用户E；归属于IDC3的用户有用户F和用户G。并且用户A的流量权值为60，用户B的流量权值为70，用户C的流量权值为55，用户D的流量权值为40，用户E的流量权值为80，用户F的流量权值为65，用户G的流量权值为90。则根据上述信息确定IDC1的高峰流量占比为：(60+70)/(60+70+55+40+80+65+90)＝28％；确定IDC2的高峰流量占比为：(55+40+80)/(60+70+55+40+80+65+90)＝38％；确定IDC3的高峰流量占比为：(65+90)/(60+70+55+40+80+65+90)＝34％。
本公开内容实施例全局所有的IDC包括但不限于第一IDC和第二IDC两大类。
其中，第一IDC为根据分配流量占比和高峰流量占比，确定高峰流量占比大于分配流量占比的IDC，即在繁忙时期，第一IDC能够提供的流量无法满足归属于该IDC的用户需要使用的流量，第一IDC为需要进行用户转移的IDC。
第二IDC为根据分配流量占比和高峰流量占比，确定高峰流量占比小于分配流量占比的IDC，即在繁忙时期，第二IDC能够提供的流量大于归属于该IDC的用户需要使用的流量，第二IDC为能够接纳用户的IDC。
例如，假设全局共包括两个IDC，分别为IDC1和IDC2；IDC1的分配流量占比为50％，IDC2的分配流量占比为50％。全局用户包括用户A、用户B、用户C、用户D、和用户F，其中，用户与用户归属的IDC的映射关系如图4所示，归属于IDC1的用户有用户A、用户B和用户E；归属于IDC2的用户有用户C、用户D和用户F；并且用户A的流量权值为60，用户B的流量权值为70，用户C的流量权值为55，用户D的流量权值为40，用户E的流量权值为80，用户F的流量权值为65。则根据上述信息确定IDC1的高峰流量占比为：(60+70+80)/(60+70+55+40+80+65)＝57％；确定IDC2的高峰流量占比 为：(55+40+65)/(60+70+55+40+80+65)＝43％。
根据DC1、IDC2和IDC3的分配流量占比和高峰流量占比，确定IDC1的高峰流量占比大于分配流量占比，则确定IDC1为第一IDC，即需要将IDC1的部分用户进行转移；确定IDC2的高峰流量占比小于分配流量占比，则确定IDC2为第二IDC，即IDC2能够接纳IDC1中需要转移的用户。
本公开内容实施例在确定需要进行用户转移的第一IDC和能够接纳用户的第二IDC之后，还需要确定第一IDC中需要转移的用户。由于针对每一个第一IDC，确定第一IDC中需要转移的用户的方法相同，下面以一个第一IDC为例进行说明。
可选的，本公开内容实施例在将第一数据中心中需要转移的用户转移处理到确定的第二数据中心时，具体可以采用下列方式：
针对任意一个第一IDC，根据IDC和用户的绑定关系，确定归属所述第一IDC的所有用户；
从归属所述第一IDC的所述所有用户中，确定所述第一IDC中需要进行转移的用户，以及从所有所述能够接纳用户的第二IDC中，确定能够接纳所述第一IDC中所述需要进行转移的用户的至少一个所述第二IDC；并将所述需要进行转移的用户转移处理到所述确定的至少一个所述第二IDC。
本公开内容实施例针对任意一个第一IDC，根据下列步骤确定第一IDC中需要进行转移的用户。
1、根据IDC和用户的绑定关系，确定归属所述第一IDC的所有用户；
本公开内容实施例可以根据路由表确定归属于该第一IDC的所有用户。例如，本公开内容实施例的路由表如图5所示，即：Set<<用户A，IDC1>，<用户B，IDC1>，<用户C，IDC2>，<用户D，IDC2>，<用户E，IDC2>，<用户F，IDC3>>，并且若IDC2为第一IDC，则在确定第一IDC对应的所有用户时，根据路由表，确定归属IDC2的所有用户为用户D、用户E和用户F。
2、从归属所述第一IDC的所有用户中，确定所述第一IDC中需要进行转 移的用户。
具体的，确定归属所述第一IDC的每一个用户的用户流量占比；根据所述用户流量占比，从归属所述第一IDC的所有用户中选取需要进行转移的用户，其中选取的所述用户的用户流量占比之和不小于所述第一IDC的所述高峰流量占比与所述分配流量占比之差。
本公开内容实施例用户的用户流量占比为用户的流量权值与全局所有用户流量权值之和的比值。
例如，全局所有用户包括用户A、用户B、用户C、用户D、用户E和用户F，并且用户A的流量权值为60，用户B的流量权值为70，用户C的流量权值为50，用户D的流量权值为40，用户E的流量权值为80，用户F的流量权值为90，则确定用户A的用户流量占比为：60/(60+70+50+50+80+90)＝15％；确定用户B的用户流量占比为：70/(60+70+50+50+80+90)＝17.5％；确定用户C的用户流量占比为：50/(60+70+50+50+80+90)＝12.5％；确定用户D的用户流量占比为：50/(60+70+50+50+80+90)＝12.5％；确定用户E的用户流量占比为：80/(60+70+50+50+80+90)＝20％；确定用户F的用户流量占比为：90/(60+70+50+50+80+90)＝25.5％。
本公开内容实施例首先获取归属该第一IDC的所有用户的用户流量占比，从该第一IDC的所有用户中选取出需要进行转移的用户，其中该需要进行转移的用户的用户流量占比之和不小于该第一IDC的高峰流量占比与分配流量占比之差。
本公开内容实施例在从该第一IDC的所有用户中选取出需要进行转移的用户时，可以采用下面两种方式选取。
方式一、本公开内容实施例根据每个用户的用户流量占比，随机选取出M个用户，并且该M个用户的用户流量占比之和不小于该第一IDC的高峰流量占比与分配流量占比之差。
例如，IDC1为第一IDC，归属IDC1的用户有用户A、用户B、用户C、 用户D、用户E和用户F和用户G，并且用户A的用户流量占比为5％，用户B的用户流量占比为2％，用户C的用户流量占比为10％，用户D的用户流量占比为6％，用户E的用户流量占比为9％，用户F的用户流量占比为3％；假设IDC1的高峰流量占比与分配流量占比之差8％，则从用户A、用户B、用户C、用户D、用户E、用户F和用户G中，根据其对应的用户流量占比，选取用户A和用户F作为需要进行转移的用户。
方式二、根据所述用户流量占比，将从高到低排序的前N个用户作为所述需要进行转移的用户；其中N为正整数。
并且选取的N个用户的用户流量占比之和不小于所述第一IDC的高峰流量占比与分配流量占比之差。
本公开内容实施例在采用方式二确定需要转移的用户时，需要将所述第一IDC的所有用户根据用户流量占比从大到小进行排序，在需要确定需要转移的用户时，按照从大到小的顺序选取出N个用户，并且该N个用户的用户流量占比之和不小于所述第一IDC的高峰流量占比与分配流量占比之差。
例如，IDC1为第一IDC，归属IDC1的用户有用户A、用户B、用户C、用户D、用户E、用户F和用户G，并且用户A的用户流量占比为5％，用户B的用户流量占比为2％，用户C的用户流量占比为10％，用户D的用户流量占比为6％，用户E的用户流量占比为9％，用户F的用户流量占比为3％；假设DC1的高峰流量占比与分配流量占比之差13％。在确定需要转移的用户时，先将归属IDC1所有用户根据用户流量占比从大到小排列，依次为用户C、用户E、用户D、用户A、用户F和用户B，则从归属IDC1的用户中选取两个用户，即用户C和用户E，将用户C和用户E作为需要进行转移的用户。
本公开内容实施例在确定第一IDC中需要进行转移的用户之后，还需要确定能够接纳该第一IDC中需要转移的用户的第二IDC。
针对一个第一IDC，在确定该第一IDC中需要进行转移的用户之后。需要为该IDC中需要转移的用户分配第二IDC，即从所有第二IDC中确定能够接 纳所述第一IDC中需要进行转移的用户的至少一个第二IDC。
具体的，根据设定的转移原则，从所有所述能够接纳用户的第二IDC中，确定能够接纳所述第一IDC中所述需要进行转移的用户的至少一个第二IDC。
下面对本公开内容实施例设定的转移原则进行详细说明。
本公开内容实施例设定的转移原则包括下面转移原则中一个。
转移原则一、根据所有所述第一IDC的第一差值，将所有所述第一IDC按照从大到小进行排序；以及根据所有所述第二IDC的第二差值，将所有所述第二IDC按照从大到小进行排序；
针对任意一个所述第一IDC，确定与所述第一IDC匹配的所述第二IDC，其中与所述第一IDC匹配的所述第二IDC在排序后的所有所述第二IDC中的位置不低于与目标第一IDC匹配的所述第二IDC在排序后的所有所述第二IDC中的位置，所述目标第一IDC在排序后的所有所述第一IDC中的位置低于所述第一IDC在排序后的所有所述第一IDC中的位置；
其中，所述第一IDC的第一差值为所述第一IDC的高峰流量占比和所述第一IDC的分配流量占比的差值；所述第二IDC的第二差值为所述第二IDC的分配流量占比和所述第二IDC的高峰流量占比的差值。
实施中，根据本公开内容实施例提供的转移原则一，在将第一IDC中需要转移的用户转移到能够接纳用户的第二IDC之前，需要根据每一个第一IDC的第一差值，将每一个第一IDC根据第一差值的大小，由大到小进行排列；并且，根据每一个第二IDC，将每一个第二IDC根据第二差值大小，由大到小进行排序；在将所有第一IDC和所有第二IDC进行排序之后，根据第一IDC的排序，将第一IDC中需要转移的用户依次转移，并且，将第一IDC中需要转移的用户优先转移到排序在前的第二IDC中。
例如，第一IDC包括IDC1、IDC2、IDC3和IDC4，第二IDC包括IDC5、IDC6和IDC7，并且IDC1的高峰流量占比为15％，分配流量占比为10％；IDC2的高峰流量占比为21％，分配流量占比为15％；IDC3的高峰流量占比为28％， 分配流量占比为20％；IDC4的高峰流量占比为13％，分配流量占比为10％；IDC5的高峰流量占比为8％，分配流量占比为15％；IDC6的高峰流量占比为10％，分配流量占比为20％；IDC7的高峰流量占比为5％，分配流量占比为10％；则确定IDC1的第一差值为5％，IDC2的第一差值为6％，IDC3的第一差值为8％，IDC4的第一差值为3％，IDC5的第二差值为7％，IDC6的第二差值为10％，IDC7的第二差值为5％。则对所有的第一IDC进行排序为：IDC3、IDC2、IDC1、IDC4；对所有的第二IDC进行排序为：IDC6、IDC5、IDC7。则首先将IDC3中需要转移的用户流量占比之和为8％的用户转移至IDC6中；再将IDC2中需要转移的用户流量占比之和为2％的用户转移至IDC6中，将IDC2中需要转移的用户流量占比之和为4％的用户转移至IDC5中；再将IDC1中需要转移的用户流量占比之和为3％的用户转移至IDC5中，将IDC1中需要转移的用户流量占比之和为2％的用户转移至IDC7中；最后将IDC4中需要转移的用户流量占比之和为3％的用户转移至IDC7中，从而实现用户的转移。
转移原则二、将所述能够接纳用户的第二IDC中距离所述第一IDC最近的第二IDC，作为能够接纳所述第一IDC中所述需要进行转移的用户的第二IDC。
本公开内容实施例针对一个第一IDC，在确定该第一IDC中需要进行转移的用户后，在所有能够接纳用户的第二IDC中，确定距离需要转移的用户最近的IDC，将该第一IDC中需要进行转移的用户转移到确定的第二IDC中。
具体的，在本公开内容实施例中，针对一个第一IDC，确定需要转移的用户之后，根据需要转移的每一个用户的流量占比，根据流量占比的大小将用户从大到小进行排序，根据需要转移的用户的排序，将流量占比大的用户优先转移出去。在转移每一个用户时，需要根据用户距离能够接纳用户的第二IDC的距离，将用户转移至距离用户最近的第二IDC。
需要说明的是，本公开内容实施例在确定第一IDC的转移顺序时，可以根据所有第一IDC的差值大小，将差值较大的第一IDC中需要转移的用户优先 转移。本公开内容实施例还可以根据其他原则确定第一IDC的转移顺序。本公开内容实施例在根据转移原则二转移用户时，对第一IDC的转移顺序不作限定。
例如，第一IDC包括IDC1、IDC2、IDC3和IDC4，第二IDC包括IDC5、IDC6和IDC7，并且IDC1的高峰流量占比为15％，分配流量占比为10％；IDC2的高峰流量占比为21％，分配流量占比为15％；IDC3的高峰流量占比为28％，分配流量占比为20％；IDC4的高峰流量占比为13％，分配流量占比为10％；IDC5的高峰流量占比为8％，分配流量占比为15％；IDC6的高峰流量占比为10％，分配流量占比为20％；IDC7的高峰流量占比为5％，分配流量占比为10％；则确定IDC1的第一差值为5％，IDC2的第一差值为6％，IDC3的第一差值为8％，IDC4的第一差值为3％，IDC5的第二差值为7％，IDC6的第二差值为10％，IDC7的第二差值为5％。
在确定第一IDC的转移顺序时，可以根据第一IDC的差值大小，第一IDC按照差值从大到小的顺序排序，即：IDC3、IDC2、IDC1、IDC4。则在转移用户时，优先转移IDC3中需要转移的用户。比如，在转移IDC1中需要转移的用户时，假设IDC1中需要转移的用户为用户A、用户B和用户C，并且用户A的流量占比为4％，用户B的流量占比为3％，用户C的流量占比为1％，则在转移IDC1的用户时，优先转移用户A，并且确定能够接受用户A的第二IDC，即IDC5、IDC6和IDC7，从IDC5、IDC6和IDC7中选择距离用户A最近的IDC，假设为IDC7，则将用户A转移至IDC7中；在转移用户B时，由于IDC7只能接受1％的用户，此时能够接受用户B的第二IDC包括IDC5和IDC6，在确定距离用户B最近的IDC为IDC5后，将用户B转移至IDC5中；在转移用户C时，由于用户C的流量占比为1％，而IDC5还可以接受流量占比为4％的用户，IDC6还可以接受流量占比为10％的用户，IDC7还可以接受流量占比为1％的用户，此时确定可以接受用户C的第二IDC包括IDC5、IDC6和IDC7，并确定距离用户C最近的IDC为IDC5，则将用户C转移至IDC5中。
本公开内容实施例对确定的用户进行转移处理，其中转移处理包括但不限于：
将用户的用户数据由转移之前的IDC迁移至该用户转移后的IDC中、修改路由表中用户归属的IDC。
具体的，将用户的用户数据由转移之前的IDC迁移至该用户转移后的IDC中表示，将转移之前的IDC中记录的用户的部分或者全部用户数据迁移至用户转以后的IDC中，即将转移之前的IDC中记录的关于该用户的用户数据删除。
例如，本公开内容实施例的IDC包括IDC1、IDC2、IDC3和IDC4，用户包括用户A、用户B、用户C、用户D，用户E、用户F、用户G、用户H、用户I和用户J，其中，A、B、C、D、E、F、G、H、I和J为用户ID。在根据IDC的高峰流量占比和分配流量占比将用户转移之前，用户和IDC的映射关系如图6A所示，用户A、用户B和用户E归属于IDC1；用户C、用户F和用户G归属于IDC2；用户D和用户H归属于IDC3；用户I和用户J归属于IDC4。则所有用户与用户归属的IDC的映射关系构成的集合，采用路由表的形式表示，如图6B所示，为：Set<<用户A，IDC1>，<用户B，IDC1>，<用户C，IDC2>，<用户D，IDC3>，<用户E，IDC1>，<用户F，IDC2>，<用户G，IDC2>，<用户H，IDC3>，<用户I，IDC4>，<用户J，IDC4>>。
在确定每一个IDC的高峰流量占比和分配流量占比之后，确定IDC1和IDC2为第一IDC，IDC3和IDC4为第二IDC，并且需要将IDC1中的用户A转移至IDC3中，将IDC2中的用户F转移至IDC4中，则在将用户A和用户F进行转移后,用户和IDC的映射关系如图6C所示，用户B和用户E归属于IDC1；用户C和用户G归属于IDC2；用户A、用户D和用户H归属于IDC3；用户F、用户I和用户J归属于IDC4。修改路由表中用户归属的IDC，修改后的路由表如图6D所示，为：Set<<用户A，IDC3>，<用户B，IDC1>，<用户C，IDC2>，<用户D，IDC3>，<用户E，IDC1>，<用户F，IDC4>，<用户G，IDC2>，<用户H，IDC3>，<用户I，IDC4>，<用户J，IDC4>>。
如图7所示，本公开内容实施例为用户分配IDC方法的整体流程图。
步骤701、确定每个IDC的分配流量占比，以及确定每个IDC的高峰流量占比；
步骤702、确定需要进行用户转移的第一IDC和能够接纳用户的第二IDC，其中所述第一IDC为高峰流量占比大于分配流量占比，第二IDC为高峰流量占比小于分配流量占比；
步骤703、针对任意一个第一IDC，根据IDC和用户的绑定关系，确定归属所述第一IDC的所有用户；
步骤704、确定归属所述第一IDC的所有用户的用户流量占比；
步骤705、根据所述用户流量占比，将从高到低排序的前N个用户作为所述需要进行转移的用户，其中选取的用户的用户流量占比之和不小于所述第一IDC的高峰流量占比与分配流量占比之差；
步骤706、根据所有第一IDC的第一差值，将所有第一IDC按照从大到小进行排序；以及根据所有第二IDC的第二差值，将所有第二IDC按照从大到小进行排序；针对任意一个第一IDC，确定与所述第一IDC匹配的第二IDC；其中与第一IDC匹配的第二IDC在排序后的所有第二IDC中的位置不低于与目标第一IDC匹配的第二IDC在排序后的所有第二IDC中的位置，所述目标第一IDC在排序后的所有所述第一IDC中的位置低于所述第一IDC在排序后的所有所述第一IDC中的位置；
步骤707、将能够接纳用户的第二IDC中距离所述第一IDC最近的第二IDC，作为能够接纳第一IDC中所述需要进行转移的用户的第二IDC；
步骤708、对确定的所述用户进行转移处理。
可选的，本公开内容实施例为用户分配IDC的时段包括繁忙时期和非繁忙时期。
其中，繁忙时期为可预期的流量高峰时期；非繁忙时期为日常情况下，除了繁忙时期之外的时段。比如，阿里巴巴集团的全球双十一购物狂欢节，即在 11月11日当天，访问阿里巴巴互联的用户突然增多，每个用户产生的流量权值突然增大，则该时期即为繁忙时期。
可选的，本公开内容实施例在繁忙时期可以采用上述为用户分配IDC的方法，根据每个IDC的高峰流量占比和分配流量占比，确定出第一IDC和第二IDC，实现将第一IDC的部分用户转移至第二IDC中。为用户合理地分配IDC，充分满足在繁忙时期用户的需要。
本公开内容实施例在非繁忙时段，为用户就近分配IDC。
由于在非繁忙时期，针对每一个IDC，归属于该IDC的用户产生的流量权值较小，在正常情况下，每一个IDC实际产生的流量占比远远小于该IDC的分配流量占比，每一个IDC均能满足归属于该IDC的用户的需要，为了提高用户体验，为用户就近分配IDC。
下面具体说明为用户就近分配IDC的方法。
具体的，本公开内容实施例在为用户分配IDC时，根据用户的ID确定用户连续登录的归属地，根据用户连续登录的归属地确定用户对应的IDC，并且将用户与用户对应的IDC的关系采用用户ID与IDC的映射关系表示。
其中，本公开内容实施例用户的ID可以是用户在该互联网上注册时，为用户生成的唯一身份标识ID。
如图8所示，本公开内容实施例一种生成IDC和用户的绑定关系的方法，包括：
步骤801、根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地；
步骤802、从所述多次确定的所述用户登录时的归属地中确定同一个归属地出现的次数；
步骤803、根据次数最多的归属地，确定所述用户绑定的IDC，并生成所述数据中心和用户的绑定关系。
本公开内容实施例在确定该用户归属的IDC时，首先确定用户的归属地。 本公开内容实施例采用下列方法确定用户的归属地。
本公开内容实施例获取用户的历史访问日志，采用DNS技术，解析出用户历史登录的归属地，通过大数据分析，确定用户的归属地。
具体的，可以获取用户连续一个月的所有访问日志，基于DNS解析出的所有访问日志对应的归属地，统计出每一个归属地被解析出的次数，将出现次数最多的归属地作为用户的归属地。
例如，针对用户A，获取用户A连续一个月的访问日志，对用户A连续一个月的访问日志采用DNS解析，确定用户A连续一个月访问日志分别对应的归属地，以及每个归属地被解析出的次数。若解析出用户A一个月内登录时的归属地有北京和杭州，并且根据访问日志，确定用户A在北京登录25次，杭州登录3次，则确定用户A的归属地为北京。
本公开内容实施例在确定用户的归属地之后，根据用户的归属地确定用户绑定的IDC。具体的，本公开内容实施例将距离用户的归属地最近的IDC作为用户绑定的IDC。
例如，针对用户A，获取用户A连续一个月的访问日志，对用户A连续一个月的访问日志采用DNS解析，确定用户A连续一个月访问日志分别对应的归属地，以及每个归属地被解析出的次数。若解析出用户A一个月内登录时的归属地有北京和杭州，并且根据访问日志，确定用户A在北京登录25次，杭州登录3次，则确定用户A的归属地为北京。并且距离北京最近的IDC为IDC2，则确定用户A对应的IDC为IDC2。
本公开内容实施例根据上述方式确定每一个用户对应的IDC之后，将每一个用户与其归属的IDC的映射关系作为一个集合，该集合为路由表。例如，假设用户A对应的IDC为IDC1，用户B对应的IDC为IDC1，用户C对应的IDC为IDC2，用户D对应的IDC为IDC3，用户E对应的IDC为IDC3，用户F对应的IDC为IDC4，则采用路由表的形式表示为：Set<<用户A，IDC1>，<用户B，IDC1>，<用户C，IDC2>，<用户D，IDC3>，<用户E，IDC3>，<用 户F，IDC4>>。本公开内容实施例在用户每次登录时，均通过DNS解析，根据用户的IP地址确定用户登录时的归属地，并确定用户登录时归属地对应的IDC，比较用户登录时的归属地对应的IDC和用户当前绑定的IDC，根据比较结果确定IDC为用户提供服务的方式。
一、用户登录时的归属地对应的IDC和用户绑定的IDC相同，通过用户登录时的归属地对应的IDC为用户提供网络服务。
例如，针对用户B，在用户B登录时，通过DNS技术，确定用户B当前登录的归属地为杭州，确定杭州对应的IDC为IDC3；而在路由表中，用户B当前绑定的IDC为IDC3，则确定用户B登录时归属地杭州对应的IDC与用户当前绑定的IDC为同一个IDC，则直接通过IDC3为用户提供网络服务。
二、用户登录时的归属地对应的IDC和用户当前绑定的IDC不同，则通过用户当前绑定的IDC为用户提供服务。
具体的，若所述用户登录时的归属地与所述用户当前绑定的IDC对应的归属地不同，控制所述当前绑定的IDC为所述用户提供网络服务。
例如，针对用户B，在用户B登录时，通过DNS技术，确定用户B当前登录的归属地为杭州，确定杭州对应的IDC为IDC3；而在路由表中，用户B当前绑定的IDC为IDC4，则确定用户B登录时归属地杭州对应的IDC与用户当前绑定的IDC不是同一个IDC，则通过用户B当前绑定的IDC4为用户提供网络服务。
如图9所示，本公开内容实施例还提供一种IDC为用户提供服务的方法，包括：
步骤901、在确定用户登录时，获取用户的访问日志；
步骤902、采用DNS技术，根据用户IP地址确定用户登录时的归属地；
步骤903、调用路由表，根据路由表确定用户当前绑定的IDC；
步骤904、判断用户登录时的归属地与用户当前绑定的IDC对应的归属地是否相同；若是，执行步骤905，若否，执行步骤906；
步骤905、控制用户登录时的归属地对应的IDC为用户提供网络服务；
步骤906、跳转至用户当前绑定的IDC，控制用户当前绑定的IDC为用户提供网络服务。
由于本公开内容实施例在用户每次登录时，首先根据用户的访问日志，确定用户登录时的归属地对应的IDC，若在确定用户登录时的归属地对应的IDC与用户当前绑定的IDC不同时，跳转至用户当前绑定的IDC，控制用户当前绑定的IDC为用户提供网络服务。但是，若在设定的时长内，确定用户每次登录时的归属地对应的IDC均为同一个，并且与用户当前绑定的IDC不同，则需要修改用户当前绑定的IDC。
具体的，若根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地对应的IDC在设定时长内为同一个，且与所述用户当前绑定的IDC不同，则根据所述用户登录时的归属地对应的IDC，修改所述用户当前绑定的IDC。
本公开内容实施例在用户每次登录时，首先根据用户的访问日志，采用DNS解析，根据用户的IP地址确定用户的归属地，确定用户归属地对应的的IDC，其中，用户归属地对应的的IDC为距离用户登录的归属地最近的IDC。在确定用户最近的IDC后，还需要调用表示IDC与用户绑定关系的路由表，通过调用路由表确定距离用户最近的IDC与路由表中用户当前绑定的IDC是否相同。若距离用户最近的IDC与路由表中用户当前绑定的IDC不是同一个IDC，并且在设定的时长内，用户每次登录时最近的IDC均为同一个，并且与用户当前绑定的IDC不是同一个IDC，则确定用户归属地发生变化，此时，需要修改用户当前绑定的IDC。
其中，判断修改用户当前绑定的IDC的条件为：
在非繁忙时段的设定时长内，若根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地对应的IDC为同一个，且所述用户登录时的所述归属地对应的IDC与所述用户当前绑定的IDC不同，则将所述用户登录时的所 述归属地对应的所述数据中心作为所述用户绑定的数据中心。
本公开内容实施例的设定时长可以是30天，或者50天。本公开内容实施例对设定时长不作限定。
例如，用户A为阿里巴巴中国用户，并且用户A当前绑定的IDC为IDC1，其中IDC1为位于中国的服务器；在用户A移民美国后，在用户A在美国登录阿里巴巴网站时，根据用户A的访问日志，根据用户A的IP地址确定距离用户A最近的IDC为美国的IDC5，在确定距离用户A最近的IDC与用户A当前绑定的IDC不同时，跳转至IDC1为用户A提供网络服务。但是在连续30天，每次用户A登录阿里巴巴网站时，确定距离A最近的IDC均为IDC5，则确定将用户A当前绑定的IDC修改为IDC5。
可选的，若所述IDC和用户的绑定关系发生变化后，将所述用户转移处理到所述变化后的数据中心。
本公开内容实施例对用户进行转移处理包括将用户的部分或全部用户数据进行迁移，即将用户的部分或全部用户数据由原来绑定的IDC迁移至修改后绑定的IDC中。
需要说明的是，本公开内容实施例在其确定IDC和归属IDC的用户的绑定关系发生变化后，可以首先修改路由表中用户对应的归属IDC，然后将用户的部分或全部用户数据迁移至修改后的IDC；还可以是首先将用户的部分或全部用户数据迁移至修改后的IDC，然后修改路由表中用户对应的归属IDC。本公开内容实施例对修改用户当前绑定的IDC和对用户进行转移处理的先后顺序不作限定。
例如，用户A为阿里巴巴中国用户，并且用户A当前绑定的IDC为IDC1，其中IDC1为位于中国的服务器；在用户A移民美国后，在用户A在美国登录阿里巴巴网站时，根据用户A的访问日志，根据用户A的IP地址确定距离用户A最近的IDC为美国的IDC5，在确定距离用户A最近的IDC与用户A当前绑定的IDC不同时，跳转至IDC1为用户A提供网络服务。但是在连续30 天，每次用户A登录阿里巴巴网站时，确定距离A最近的IDC均为IDC5，则确定将用户A当前绑定的IDC修改为IDC5，并将用户A在IDC1中的部分或全部用户数据迁移至IDC5中。
需要说明的是，本公开内容实施例在繁忙时段，每一个用户初始归属的IDC可以采用本公开内容实施例提供的在非繁忙时段为用户分配IDC的方法确定，也可以采用其他方法确定每一个用户初始归属的IDC。
基于同一发明构思，本公开内容实施例中还提供了一种为用户分配数据中心的设备，由于该设备解决问题的原理与本公开内容实施例为用户分配数据中心的方法相似，因此该设备的实施可以参见方法的实施，重复之处不再赘述。
如图10所示，本公开内容实施例一种为用户分配数据中心的设备，包括：
获取模块1001，用于确定每个数据中心的分配流量占比，以及确定所述每个数据中心的高峰流量占比；
确定模块1002，用于确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心；
处理模块1003，用于将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心；其中所述第一数据中心为所述高峰流量占比大于所述分配流量占比的数据中心，所述第二数据中心为所述高峰流量占比小于所述分配流量占比的数据中心。
可选的，所述数据中心的分配流量占比为所述数据中心承载用户流量的上限与全部数据中心承载用户流量的上限之和的比值。
可选的，所述获取模块1001，具体用于：
针对任意一个所述数据中心，将归属所述数据中心的用户的流量权值之和，与归属全部数据中心的所有用户的所述流量权值之和的比值，作为所述数据中心的高峰流量占比。
可选的，所述用户的流量权值为表示所述用户产生的用户流量占全部数据中心的所有用户产生的用户流量之和的比重的数值。
可选的，所述处理模块1003，具体用于：
针对任意一个所述第一数据中心，根据数据中心和用户的绑定关系，确定归属所述第一数据中心的所有用户；从归属所述第一数据中心的所述所有用户中，确定所述第一数据中心中需要进行转移的用户，以及从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个所述第二数据中心；并将所述需要进行转移的用户转移处理到所述确定的至少一个所述第二数据中心。
可选的，所述处理模块1003，具体用于：
确定归属所述第一数据中心的每一个用户的用户流量占比；根据所述用户流量占比，从归属所述第一数据中心的所有所述用户中选取需要进行转移的用户，其中选取的所述用户的所述用户流量占比之和不小于所述第一数据中心的所述高峰流量占比与所述分配流量占比之差。
可选的，所述处理模块1003，具体用于：
根据所述用户流量占比，将从高到低排序的前N个用户作为所述需要进行转移的用户；其中N为正整数。
可选的，所述处理模块1003，具体用于：
根据设定的转移原则，从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个第二数据中心；
所述转移原则包括下列转移原则中的一个：
转移原则一、根据所有所述第一数据中心的第一差值，将所有所述第一数据中心按照从大到小进行排序；以及根据所有所述第二数据中心的第二差值，将所有所述第二数据中心按照从大到小进行排序；针对任意一个所述第一数据中心，确定与所述第一数据中心匹配的所述第二数据中心，其中与所述第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置不低于与目标第一数据中心匹配的所述第二数据中心在排序后的所有所述第 二数据中心中的位置，所述目标第一数据中心在排序后的所有所述第一数据中心中的位置低于所述第一数据中心在排序后的所有所述第一数据中心中的位置；其中，所述第一数据中心的第一差值为所述第一数据中心的高峰流量占比和所述第一数据中心的分配流量占比的差值；所述第二数据中心的第二差值为所述第二数据中心的分配流量占比和所述第二数据中心的高峰流量占比的差值；
转移原则二、将所述能够接纳用户的第二数据中心中距离所述第一数据中心最近的第二数据中心，作为能够接纳所述第一数据中心中所述需要进行转移的用户的第二数据中心。
可选的，所述处理模块1003，还用于，根据下列方式生成所述数据中心和用户的绑定关系：
根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地；从所述多次确定的所述用户登录时的归属地中确定同一个归属地出现的次数；根据次数最多的归属地，确定所述用户绑定的数据中心，并生成所述数据中心和用户的绑定关系。
可选的，所述确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心的步骤是在繁忙时段内完成的；
所述根据不同时间的用户的访问日志，多次确定所述用户的归属地的步骤是在非繁忙时段内完成的。
可选的，所述处理模块1003，还用于：
在非繁忙时段的设定时长内，若根据不同时间的所述用户的访问日志，多次确定所述用户登录时的归属地对应的数据中心为同一个，且所述用户登录时的所述归属地对应的所述数据中心与所述用户当前绑定的数据中心不同，则将所述用户登录时的所述归属地对应的所述数据中心作为所述用户绑定的数据中心。
可选的，所述处理模块1003，还用于：
在非繁忙时段内，若所述数据中心和用户的绑定关系发生变化后，将所述用户转移处理到所述变化后的数据中心。
可选的，所述处理模块1003，还用于：
在非繁忙时段内，若所述用户登录时的归属地与所述用户当前绑定的数据中心对应的归属地不同，控制所述当前绑定的数据中心为所述用户提供网络服务。
以上参照示出根据本公开内容实施例的方法、装置(系统)和/或计算机程序产品的框图和/或流程图描述本公开内容。应理解，可以通过计算机程序指令来实现框图和/或流程图示图的一个块以及框图和/或流程图示图的块的组合。可以将这些计算机程序指令提供给通用计算机、专用计算机的处理器和/或其它可编程数据处理装置，以产生机器，使得经由计算机处理器和/或其它可编程数据处理装置执行的指令创建用于实现框图和/或流程图块中所指定的功能/动作的方法。
相应地，还可以用硬件和/或软件(包括固件、驻留软件、微码等)来实施本公开内容。更进一步地，本公开内容可以采取计算机可使用或计算机可读存储介质上的计算机程序产品的形式，其具有在介质中实现的计算机可使用或计算机可读程序代码，以由指令执行系统来使用或结合指令执行系统而使用。在本公开内容上下文中，计算机可使用或计算机可读介质可以是任意介质，其可以包含、存储、通信、传输、或传送程序，以由指令执行系统、装置或设备使用，或结合指令执行系统、装置或设备使用。
显然，本领域的技术人员可以对本公开内容进行各种改动和变型而不脱离本公开内容的精神和范围。这样，倘若本公开内容的这些修改和变型属于本公开内容权利要求及其等同技术的范围之内，则本公开内容也意图包含这些改动和变型在内。

Claims (26)

  1. 一种为用户分配数据中心的方法，其特征在于，该方法包括：
    确定每个数据中心的分配流量占比，以及确定所述每个数据中心的高峰流量占比；
    确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心；
    将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心；
    其中所述第一数据中心为所述高峰流量占比大于所述分配流量占比的数据中心，所述第二数据中心为所述高峰流量占比小于所述分配流量占比的数据中心。
  2. 如权利要求1所述的方法，其特征在于，所述数据中心的分配流量占比为所述数据中心承载用户流量的上限与全部数据中心承载用户流量的上限之和的比值。
  3. 如权利要求1所述的方法，其特征在于，所述确定所述每个数据中心的高峰流量占比，包括：
    针对任意一个所述数据中心，将归属所述数据中心的用户的流量权值之和，与归属全部数据中心的所有用户的所述流量权值之和的比值，作为所述数据中心的高峰流量占比。
  4. 如权利要求3所述的方法，其特征在于，所述用户的流量权值为表示所述用户产生的用户流量占全部数据中心的所有用户产生的用户流量之和的比重的数值。
  5. 如权利要求1所述的方法，其特征在于，所述将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心，包括：
    针对任意一个所述第一数据中心，根据数据中心和用户的绑定关系，确定归属所述第一数据中心的所有用户；
    从归属所述第一数据中心的所述所有用户中，确定所述第一数据中心中需要进行转移的用户，以及从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一 个所述第二数据中心；并将所述需要进行转移的用户转移处理到所述确定的至少一个所述第二数据中心。
  6. 如权利要求5所述的方法，其特征在于，所述确定所述第一数据中心中需要进行转移的用户，包括：
    确定归属所述第一数据中心的每一个用户的用户流量占比；
    根据所述用户流量占比，从归属所述第一数据中心的所有所述用户中选取需要进行转移的用户，其中选取的所述用户的所述用户流量占比之和不小于所述第一数据中心的所述高峰流量占比与所述分配流量占比之差。
  7. 如权利要求6所述的方法，其特征在于，所述根据所述用户流量占比，从归属所述第一数据中心的所有所述用户中选取需要进行转移的用户，包括：
    根据所述用户流量占比，将从高到低排序的前N个用户作为所述需要进行转移的用户；
    其中N为正整数。
  8. 如权利要求5所述的方法，其特征在于，所述从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个所述第二数据中心，包括：
    根据设定的转移原则，从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个所述第二数据中心；
    所述转移原则包括下列转移原则中的一个：
    转移原则一、根据所有所述第一数据中心的第一差值，将所有所述第一数据中心按照从大到小进行排序；以及根据所有所述第二数据中心的第二差值，将所有所述第二数据中心按照从大到小进行排序；
    针对任意一个所述第一数据中心，确定与所述第一数据中心匹配的所述第二数据中心，其中与所述第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置不低于与目标第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置，所述目标第一数据中心在排序后的所有所述第一数据中心中的位置低于所述第一数据中心在排序后的所有所述第一数据中心中的位置；
    其中，所述第一数据中心的第一差值为所述第一数据中心的高峰流量占比和所述第一数据中心的分配流量占比的差值；所述第二数据中心的第二差值为所述第二数据中心的分配流量占比和所述第二数据中心的高峰流量占比的差值；
    转移原则二、将所述能够接纳用户的第二数据中心中距离所述第一数据中心最近的第二数据中心，作为能够接纳所述第一数据中心中所述需要进行转移的用户的第二数据中心。
  9. 如权利要求5所述的方法，其特征在于，根据下列方式生成所述数据中心和用户的绑定关系：
    根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地；
    从所述多次确定的所述用户登录时的归属地中确定同一个归属地出现的次数；
    根据次数最多的归属地，确定所述用户绑定的数据中心，并生成所述数据中心和用户的绑定关系。
  10. 如权利要求9所述的方法，其特征在于，所述确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心的步骤是在繁忙时段内完成的；
    所述根据不同时间的用户的访问日志，多次确定所述用户的归属地的步骤是在非繁忙时段内完成的。
  11. 如权利要求10所述的方法，其特征在于，在所述根据次数最多的归属地，确定所述用户绑定的数据中心的步骤之后，还包括：
    在非繁忙时段的设定时长内，若根据不同时间的所述用户的访问日志，多次确定所述用户登录时的归属地对应的数据中心为同一个，且所述用户登录时的所述归属地对应的所述数据中心与所述用户当前绑定的数据中心不同，则将所述用户登录时的所述归属地对应的所述数据中心作为所述用户绑定的数据中心。
  12. 如权利要求10或11所述的方法，其特征在于，该方法还包括：
    在非繁忙时段内，若所述数据中心和用户的绑定关系发生变化后，将所述用户转移处理到所述变化后的数据中心。
  13. 如权利要求9～11任一所述的方法，其特征在于，该方法还包括：
    在非繁忙时段内，若所述用户登录时的归属地与所述用户当前绑定的数据中心对应的归属地不同，控制所述当前绑定的数据中心为所述用户提供网络服务。
  14. 一种为用户分配数据中心的设备，其特征在于，包括：
    获取模块，用于确定每个数据中心的分配流量占比，以及确定所述每个数据中心的高峰流量占比；
    确定模块，用于确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心；
    处理模块，用于将所述第一数据中心中需要转移的用户转移处理到所述确定的第二数据中心；其中所述第一数据中心为所述高峰流量占比大于所述分配流量占比的数据中心，所述第二数据中心为所述高峰流量占比小于所述分配流量占比的数据中心。
  15. 如权利要求14所述的设备，其特征在于，所述数据中心的分配流量占比为所述数据中心承载用户流量的上限与全部数据中心承载用户流量的上限之和的比值。
  16. 如权利要求14所述的设备，其特征在于，所述获取模块，具体用于：
    针对任意一个所述数据中心，将归属所述数据中心的用户的流量权值之和，与归属全部数据中心的所有用户的所述流量权值之和的比值，作为所述数据中心的高峰流量占比。
  17. 如权利要求16所述的设备，其特征在于，所述用户的流量权值为表示所述用户产生的用户流量占全部数据中心的所有用户产生的用户流量之和的比重的数值。
  18. 如权利要求14所述的设备，其特征在于，所述处理模块，具体用于：
    针对任意一个所述第一数据中心，根据数据中心和用户的绑定关系，确定归属所述第一数据中心的所有用户；从归属所述第一数据中心的所述所有用户中，确定所述第一数据中心中需要进行转移的用户，以及从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中 所述需要进行转移的用户的至少一个所述第二数据中心；并将所述需要进行转移的用户转移处理到所述确定的至少一个所述第二数据中心。
  19. 如权利要求18所述的设备，其特征在于，所述处理模块，具体用于：
    确定归属所述第一数据中心的每一个用户的用户流量占比；根据所述用户流量占比，从归属所述第一数据中心的所有所述用户中选取需要进行转移的用户，其中选取的所述用户的所述用户流量占比之和不小于所述第一数据中心的所述高峰流量占比与所述分配流量占比之差。
  20. 如权利要求19所述的设备，其特征在于，所述处理模块，具体用于：
    根据所述用户流量占比，将从高到低排序的前N个用户作为所述需要进行转移的用户；其中N为正整数。
  21. 如权利要求18所述的设备，其特征在于，所述处理模块，具体用于：
    根据设定的转移原则，从所有所述能够接纳用户的第二数据中心中，确定能够接纳所述第一数据中心中所述需要进行转移的用户的至少一个第二数据中心；
    所述转移原则包括下列转移原则中的一个：
    转移原则一、根据所有所述第一数据中心的第一差值，将所有所述第一数据中心按照从大到小进行排序；以及根据所有所述第二数据中心的第二差值，将所有所述第二数据中心按照从大到小进行排序；针对任意一个所述第一数据中心，确定与所述第一数据中心匹配的所述第二数据中心，其中与所述第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置不低于与目标第一数据中心匹配的所述第二数据中心在排序后的所有所述第二数据中心中的位置，所述目标第一数据中心在排序后的所有所述第一数据中心中的位置低于所述第一数据中心在排序后的所有所述第一数据中心中的位置；其中，所述第一数据中心的第一差值为所述第一数据中心的高峰流量占比和所述第一数据中心的分配流量占比的差值；所述第二数据中心的第二差值为所述第二数据中心的分配流量占比和所述第二数据中心的高峰流量占比的差值；
    转移原则二、将所述能够接纳用户的第二数据中心中距离所述第一数据中心最近的第二数据中心，作为能够接纳所述第一数据中心中所述需要进行转移的用户的第二数据中心。
  22. 如权利要求18所述的设备，其特征在于，所述处理模块，还用于，根据下列方式生成所述数据中心和用户的绑定关系：
    根据不同时间的用户的访问日志，多次确定所述用户登录时的归属地；从所述多次确定的所述用户登录时的归属地中确定同一个归属地出现的次数；根据次数最多的归属地，确定所述用户绑定的数据中心，并生成所述数据中心和用户的绑定关系。
  23. 如权利要求22所述的设备，其特征在于，所述确定需要进行用户转移的第一数据中心和能够接纳用户的第二数据中心的步骤是在繁忙时段内完成的；
    所述根据不同时间的用户的访问日志，多次确定所述用户的归属地的步骤是在非繁忙时段内完成的。
  24. 如权利要求23所述的设备，其特征在于，所述处理模块，还用于：
    在非繁忙时段的设定时长内，若根据不同时间的所述用户的访问日志，多次确定所述用户登录时的归属地对应的数据中心为同一个，且所述用户登录时的所述归属地对应的所述数据中心与所述用户当前绑定的数据中心不同，则将所述用户登录时的所述归属地对应的所述数据中心作为所述用户绑定的数据中心。
  25. 如权利要求23或24所述的设备，其特征在于，所述处理模块，还用于：
    在非繁忙时段内，若所述数据中心和用户的绑定关系发生变化后，将所述用户转移处理到所述变化后的数据中心。
  26. 如权利要求22～24任一所述的设备，其特征在于，所述处理模块，还用于：
    在非繁忙时段内，若所述用户登录时的归属地与所述用户当前绑定的数据中心对应的归属地不同，控制所述当前绑定的数据中心为所述用户提供网络服务。
PCT/CN2017/078044 2016-03-31 2017-03-24 一种为用户分配数据中心的方法和设备 Ceased WO2017167129A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610201655.4A CN107295042B (zh) 2016-03-31 2016-03-31 一种为用户分配数据中心的方法和设备
CN201610201655.4 2016-03-31

Publications (1)

Publication Number Publication Date
WO2017167129A1 true WO2017167129A1 (zh) 2017-10-05

Family

ID=59962578

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/078044 Ceased WO2017167129A1 (zh) 2016-03-31 2017-03-24 一种为用户分配数据中心的方法和设备

Country Status (2)

Country Link
CN (1) CN107295042B (zh)
WO (1) WO2017167129A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109246323B (zh) * 2018-09-25 2020-09-11 国家电网有限公司客户服务中心 一种应用于异地双活话务平台的用户接入中心分配方法
CN110809051B (zh) * 2019-11-11 2020-11-13 广州华多网络科技有限公司 一种业务数据的处理方法及系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030120780A1 (en) * 2001-12-21 2003-06-26 Xiaoyun Zhu Network resource assignment system and method
US20050169254A1 (en) * 2003-04-14 2005-08-04 Fujitsu Limited Data relay apparatus, data relay method, data relay program, service selection apparatus, service selection method and service selection program
CN103259809A (zh) * 2012-02-15 2013-08-21 株式会社日立制作所 负载均衡器、负载均衡方法及分层数据中心系统
CN103874131A (zh) * 2012-12-12 2014-06-18 上海贝尔股份有限公司 一种基于服务质量的小区内负载均衡方法、装置及系统
CN103974333A (zh) * 2014-05-16 2014-08-06 西安电子科技大学 针对svc视频业务和移动速度的负载均衡方法
CN103987085A (zh) * 2013-02-07 2014-08-13 华为技术有限公司 小区间移动负荷均衡的方法和移动负荷均衡装置
CN106209745A (zh) * 2015-05-07 2016-12-07 阿里巴巴集团控股有限公司 一种流量的分流方法和设备

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6854013B2 (en) * 2001-06-25 2005-02-08 Nortel Networks Limited Method and apparatus for optimizing network service
CN101087467A (zh) * 2006-06-11 2007-12-12 上海全成通信技术有限公司 一种大容量用户并发访问的负载均衡方法
CN101009648A (zh) * 2006-12-27 2007-08-01 北京航空航天大学 多服务器的分层迭代负载平衡方法
CN101610287B (zh) * 2009-06-16 2012-03-14 浙江大学 一种应用于分布式海量存储系统的负载均衡方法
CN102469023B (zh) * 2010-11-19 2015-10-07 中国移动通信集团公司 基于云计算的调度方法、单元及系统
CN102185779B (zh) * 2011-05-11 2015-02-25 田文洪 与综合配置能力成比例的数据中心资源负载均衡的方法及装置
CN103391299B (zh) * 2012-05-08 2016-09-28 深圳市腾讯计算机系统有限公司 负载均衡方法和负载均衡系统
CN103729378A (zh) * 2012-10-16 2014-04-16 腾讯科技（深圳）有限公司 一种ip地理信息的更新方法和装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030120780A1 (en) * 2001-12-21 2003-06-26 Xiaoyun Zhu Network resource assignment system and method
US20050169254A1 (en) * 2003-04-14 2005-08-04 Fujitsu Limited Data relay apparatus, data relay method, data relay program, service selection apparatus, service selection method and service selection program
CN103259809A (zh) * 2012-02-15 2013-08-21 株式会社日立制作所 负载均衡器、负载均衡方法及分层数据中心系统
CN103874131A (zh) * 2012-12-12 2014-06-18 上海贝尔股份有限公司 一种基于服务质量的小区内负载均衡方法、装置及系统
CN103987085A (zh) * 2013-02-07 2014-08-13 华为技术有限公司 小区间移动负荷均衡的方法和移动负荷均衡装置
CN103974333A (zh) * 2014-05-16 2014-08-06 西安电子科技大学 针对svc视频业务和移动速度的负载均衡方法
CN106209745A (zh) * 2015-05-07 2016-12-07 阿里巴巴集团控股有限公司 一种流量的分流方法和设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LI, WENZHENG ET AL.: "Study and Realization of Internet Server Load Balancing", COMPUTER ENGINEERING, vol. 31, no. 6, 31 March 2005 (2005-03-31) *

Also Published As

Publication number Publication date
CN107295042B (zh) 2021-12-14
CN107295042A (zh) 2017-10-24

Similar Documents

Publication Publication Date Title
CN109040259B (zh) 一种基于mec的cdn节点分配方法和系统
US20200364608A1 (en) Communicating in a federated learning environment
CN109327542B (zh) 游戏业务访问响应方法、请求转发方法、连接方法、装置
CN107241319B (zh) 基于vpn的分布式网络爬虫系统及调度方法
CN109085999B (zh) 数据处理方法及处理系统
US20170134477A1 (en) Load balanced network file accesses
US20160308961A1 (en) Methods, Devices, and Systems for Allocating Service Nodes in a Network
CN103596066B (zh) 一种数据处理方法及装置
CN112512100B (zh) 基于切片优先级的amf重定向方法和新增管理网元
WO2015096681A1 (zh) 一种业务访问处理方法及装置
CN107317887B (zh) 一种负载均衡方法、装置和系统
EP3080695A1 (en) Allocation and migration of cloud resources in a distributed cloud system
CN113835823B (zh) 资源调度方法和装置、电子设备、计算机可读存储介质
CN104092650A (zh) 一种分配业务请求的方法和装置
CN106131244A (zh) 一种报文传送方法及装置
CN102932271A (zh) 负载均衡的实现方法和装置
CN106878052B (zh) 一种用户迁移方法和装置
CN109756584A (zh) 域名解析方法、域名解析装置及计算机可读存储介质
CN110233866A (zh) 一种负载均衡方法及负载均衡器
US20110153826A1 (en) Fault tolerant and scalable load distribution of resources
CN109413202B (zh) 区块链交易信息的排序系统及方法
CN115412737A (zh) 直播回源中继节点确定方法及装置
CN112511868B (zh) 视频播放方法、装置、电子设备及计算机存储介质
CN114039906A (zh) 流量引导方法、电子设备及可读存储介质
CN109981795A (zh) 资源请求调度方法和装置

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17773159

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17773159

Country of ref document: EP

Kind code of ref document: A1