JP6058576B2 - Slot group generation device, slot group generation method, and load distribution system - Google Patents

Description

  The present invention relates to a slot group generation device, a slot group generation method, and a load distribution system.

  In a system that accepts service requests over a network, it is necessary to flexibly change the amount of hardware to be allocated according to the amount of service requests while responding to a large number of service requests. Realization in distributed systems that operate in parallel has become the mainstream.

  In order to realize a distributed system, a load distribution device that appropriately distributes service requests and DB operations to one of the servers and equalizes the server load and the amount of stored data among the servers is required. In many Web services, since each service request can be processed independently, a load balancer is used that distributes the service requests in order or randomly, regardless of the contents of the service requests or DB operations. . However, in stateful Web services that require the same information to be shared among multiple service requests, such as SIP (Session Initiation Protocol) applications and distributed hash tables, the same service requests and DB operations are related. There is a need to implement a load balancer that distributes servers. In distributed hash tables, etc., if the distribution destination from the load balancer changes as the number of servers increases / decreases, it is necessary to move stored data, etc., but this movement generates a heavy load on the network bandwidth and storage area. For this reason, it is required to reduce the change of the distribution destination as much as possible.

  Regarding this necessity, Non-Patent Document 1 applies a consistent hash method in which a data storage server is determined according to a hash value of a key to a load balancer, and the same server as long as the same key exists in a service request. In particular, there is a technique for distributing the load uniformly while suppressing the number of slots in the slot group that defines the association between the hash value and the distribution destination server.

  Non-Patent Document 2 describes a technique for reducing the load bias of the consistent hash method by increasing the number of slots.

Michio Irie et al., “Load Balancing Method Aware of Data Replication in the Consistent Hash Method”, The Institute of Electronics, Information and Communication Engineers, IEICE Technical Report IN2010-77, October 2010 I. Stoica et al., “Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications”, Proc. ACM SIGCOMM 2001, Aug. 2001.

  In the method of Non-Patent Document 1, load increase due to the number of slot groups and server increase / decrease can be suppressed, but load equalization is not perfect, and there is a possibility that a large load deviation occurs between servers. Also, the technique of Non-Patent Document 2 shows that the load deviation is reduced by increasing the number of slots while suppressing the increase in load due to increase / decrease of the server. It is not certain that can be achieved.

  When actually building a system, in a situation where the load is not evenly distributed and it is not clear which server load will be high, select a device with sufficient performance when procuring server hardware. I must. As a result, the procurement cost for system construction will be increased. Therefore, in system construction, equalizing the load among servers in the system is a very important issue.

  Therefore, an object of the present invention is to provide a load distribution system that suppresses an increase in load due to increase / decrease in servers and does not have a load bias between servers.

  In order to solve the above-described problems and achieve the object, the present invention provides a slot group generation device that generates a slot group used for distribution of received packets in a load distribution system that distributes load among a plurality of servers. The maximum slot sequence for generating a slot sequence in which servers are arranged so that the number of appearances of permutation combinations of a predetermined number of servers is the same for each maximum server group when the number of servers constituting the load balancing system is maximized A generation unit, an extension slot sequence generation unit that generates an extended slot sequence in which one or more of the generated slot sequences are duplicated and connected, and the number of servers required for the load distribution system is the maximum server group The number of servers to be removed from the largest server group in the generated extended slot row when the number of servers is smaller than A server replacement unit that generates a slot sequence in which the number of appearances of each of the basic permutations of the non-removable servers is the same as much as possible, and a predetermined hash space A slot allocation unit that divides the slot into the slots having the length of the generated slot sequence and generates a slot group in which the servers are allocated in the order indicated by the slot sequence to each of the divided slots. And

  ADVANTAGE OF THE INVENTION According to this invention, the load distribution system which does not have the bias | biasing of the load between servers can be provided, suppressing the load increase accompanying a server increase / decrease.

FIG. 1 is a configuration diagram illustrating a load distribution system according to each embodiment. FIG. 2 is a configuration diagram illustrating the slot group generation unit. FIG. 3 is a configuration diagram illustrating the slot group generation unit. FIG. 4 is a flowchart illustrating a processing procedure of the slot group generation unit according to the first embodiment. FIG. 5 is a flowchart showing the basic slot sequence generation processing of FIG. FIG. 6 is a diagram illustrating an example of generating a basic permutation. FIG. 7 is a diagram illustrating a generation example of a basic permutation connection graph. FIG. 8 is a diagram illustrating an example of generating a basic permutation cycle. FIG. 9 is a diagram illustrating a generation example of a slot string. FIG. 10 is a diagram illustrating a generation example of a slot group in which the assigned sections of the servers A to B are assigned to the hash space. FIG. 11 is a flowchart showing a processing procedure of the slot group generation unit. FIG. 12 is a flowchart showing the processing procedure of the slot group generation unit. FIG. 13 is a diagram illustrating a generation example of the extension slot row. FIG. 14 is a diagram showing an example of a slot train when the number of servers is 3, 4, and 5. FIG. 15 is a diagram illustrating a generation example of a slot group in which the assigned sections of the servers A to E are assigned to the hash space. FIG. 16 is a flowchart illustrating a processing procedure of the maximum slot string generation unit according to the second embodiment. FIG. 17 is a diagram illustrating an example of generating extended information. FIG. 18 is a flowchart showing the slot string expansion processing of FIG. FIG. 19 is a diagram illustrating an example of generating the extended basic permutation. FIG. 20 is a diagram illustrating an example of generating a slot train after expansion. FIG. 21 is a diagram illustrating an example of updating extension information. FIG. 22 is a diagram illustrating another configuration example of the load distribution system. FIG. 23 is a diagram illustrating a computer that executes a load distribution program.

  Hereinafter, embodiments of the present invention will be described by being divided into a first embodiment and a second embodiment. In addition, this invention is not limited by each embodiment.

(First embodiment)
As shown in FIG. 1, the load distribution system (system) of each embodiment is configured by connecting a client 100, a load distribution apparatus 200, and a server 300 via a network (not shown).

  The load balancer 200 extracts a predetermined key from the packet reception unit 210 that receives a service request packet from the client 100 and the received packet, and compares the hash value with a slot group to determine a server 300 that is a packet transfer destination. A allocating unit 220 for determining the packet, a packet transferring unit 230 for performing packet transfer based on the determination of the allocating unit 220, a slot group managing unit 240 for generating a slot group for realizing uniform load, and a slot group A slot group generalization unit 250 that controls abstraction of server names and the like so that the management unit 240 can be used by many load distribution devices.

  The distribution unit 220 extracts a key extraction unit 221 from the packet received by the packet reception unit 210 as a key by the consistent hash method, generates a hash value from the extracted key, and generates a slot group generalization unit A packet transmission destination determination unit 222 that searches a slot corresponding to the hash value from the slot group acquired from 250 and determines a server assigned to the slot as a packet transfer destination.

  The slot group management unit 240 includes a list of servers 300 (the list of maximum server groups) when the maximum configuration of the system is set, a list of servers 300 to be operated (a list of requested servers 300), and what the server 300 is. Various parameters such as fault tolerance indicating whether or not the load is equalized until a table failure occurs are received, and various slot groups are generated and stored based on these parameters. The slot group management unit 240 includes a slot group generation unit 241, a slot storage unit 242, and a maximum slot sequence generation unit 243.

  The slot group generation unit 241 generates a slot group based on the parameters output from the server group abstraction unit 251. Specifically, the slot group generation unit 241 receives, as parameters from the server group abstraction unit 251, a server list (for example, A, B, C, D, E) and a basic permutation length (for example, A, B, C, D, E). 2) When the input of information such as a list of maximum server groups (for example, A, B, C, D, E) and the number of currently required servers (for example, 4) is received, the maximum slot sequence generation unit 243 receives the maximum Server group (for example, A, B, C, D, E) and the basic permutation length (for example, 2) are transmitted. Then, the slot group generation unit 241 receives, from the maximum slot sequence generation unit 243, a slot sequence that defines the slot allocation order necessary for generating a slot group with the maximum number of servers as a calculation result. Then, the slot group generation unit 241 generates a slot group based on the received slot train. Here, the slot group generation unit 241 compares the maximum number of servers and the number of servers in operation with the requested number of servers, and the requested number of servers is smaller than the maximum number of servers and the number of servers in operation. In other words, when it is necessary to reduce the number of servers in the load balancing system, a slot row (removal slot row) in which the server 300 to be removed is removed from the slot row relating to the currently operating server group. Generate. Then, a slot group is generated based on the reduced slot row, and is output to the slot storage unit 242. Details of the slot group generation unit 241 will be described later.

  The maximum slot sequence generation unit 243 generates a slot sequence of the maximum server group based on the maximum configuration server group (for example, A, B, C, D, E) and the basic permutation length (for example, 2). To do. Details of the maximum slot sequence generation unit 243 will be described later.

  The slot storage unit 242 holds the slot group and the slot string output from the slot group generation unit 241. Specifically, the slot storage unit 242 stores the slot train and the slot group generated in the process when the number of servers 300 is reduced from the maximum configuration server group to the server 300 to be operated.

  The slot group generalization unit 250 assigns a general server name to each server 300 based on the server list of the load distribution system set in advance by the system maintainer 400 or the like, and assigns the slot name to the slot group management unit 240. The server group abstraction unit 251 that requests group generation and the slot group generated by the slot group management unit 240 convert the server name assigned to each slot from a general name to a system specific name. And a server list holding table 253 that holds a correspondence table related to generalization of the server list. 1 is not limited to the number illustrated in FIG. 1, and may be an arbitrary number for the set of servers 300.

  In the following description, a case where the application service of the load distribution apparatus 200 is a service using a distributed hash table will be described as an example, but the application service is not limited to this.

  The client 100 is, for example, a DB client, and is a device that transmits a request packet related to a DB operation such as PUT or GET.

  In this case, the set of the load balancer 200 and the server 300 behaves like a database as an integrated system, the DB operation packet from the client 100 is exclusively received and distributed by the load balancer 200, and the DB operation itself is handled by the server 300. carry out. Thus, the client 100 may face the load distribution apparatus 200 without being aware of the addition of the server 300 or the like. However, since update and reference processing related to data with the same key needs to be performed consistently by the same server 300, the distribution logic of the load distribution apparatus 200 distributes data with the same key to the same server 300. There is a need. In this regard, the load balancer 200 uses a consistent hash method described below.

  In the consistent hash method, the packet reception unit 210 of the load distribution apparatus 200 that has received the DB operation packet from the client 100 passes the received DB operation packet to the key extraction unit 221 of the distribution unit 220. In order to identify operation target data, the key extraction unit 221 extracts a key from the DB operation packet and sends it to the packet transmission destination determination unit 222. The packet destination determination unit 222 that has received the key generates a hash value of the key, and obtains the hash value of the key in the slot group that is acquired from the slot group generalization unit 250 and describes the range of hash values that each server 300 is responsible for. Are determined, the server 300 that is the transfer destination of the received packet is determined. By determining the distribution destination in this way, packets can be transferred to the same server 300 as long as the keys of the operation target data are the same.

  After the packet transfer destination is determined, the packet transfer unit 230 rewrites the packet destination IP address and the like based on the information, and transfers the packet to the corresponding server 300.

  When the load distribution apparatus 200 distributes the packets by the above processing, in order to equalize the load among the servers 300, a slot group in which the range of hash values that each server 300 is in charge of is generated is generated. There is a need to. Hereinafter, a method of generating a slot group in which the range of hash values that each server 300 is in charge of, including the case where several servers 300 have failed, will be described.

  In this example, it is assumed that there are a maximum of five servers 300 (the maximum number of servers is 5) constituting the load balancing system, and the server names are DBserver1, DBserver2, DBserver3, DBserver4, and DBserver5, respectively. For example, at the beginning of operation, DBserver1, DBserver2, DBserver3, and DBserver4 will be operated on four servers 300. After that, the number of servers will be reduced to three (DBserver1, DBserver2, and DBserver3) or increased to five. It is assumed that the number of servers operated (operated) in the load balancing system increases or decreases below the maximum number of servers. In addition, regarding the response when the server 300 fails in the load balancing system, the load is made uniform by the remaining server 300 even when one of the servers 300 randomly fails.

  The slot group generalization unit 250 is responsible for name conversion so that the slot group management unit 240 can generate a slot group without being aware of the individual server name DBserver1 or the like.

  Specifically, in the server list holding table 253, the system maintainer 400 associates in advance the server name unique to the system such as DBserver1 with the server name used by the slot group management unit 240 such as A and B, for example, DBserver1: A, DBserver2: B, DBserver3: C, DBserver4: D, and DBserver5: E are held. In this situation, when the server group abstraction unit 251 receives a request from the system maintenance person 400 to generate a slot group that is resistant to one of the five servers, the server group abstraction unit In 251, a slot group for A, B, C, D, and E is required based on information in the server list holding table 253, and a slot group for A, B, C, and D is currently required. If there is, it requests the slot group generation unit 241 of the slot group management unit 240. In this case, for example, “the number of required failure tolerances (the number of assumed failure servers) +1”, that is, “1 + 1 = 2” here is transmitted as a parameter relating to slot group generation (hereinafter, this parameter is referred to as “parameter”). Called the basic permutation length). Note that the process of adding “1” is not necessarily performed by the server group abstraction unit 251 and may be performed by the slot group generation unit 241.

  In this case, for example, from the server group abstraction unit 251, the server list “A, B, C, D, E”, the basic permutation length “2”, and the maximum A, B, C, D, E When information indicating that slot groups are required and slot groups related to A, B, C, and D are currently required is transmitted to the slot group generation unit 241, the slot group management unit 240, for example, as described below A slot group composed of A, B, etc. is returned. That is, a slot group to which the assigned range of each server is assigned is returned to the circular hash space. Then, the server group individualization unit 252 converts the returned slot group from a general name such as A or B to an individual name such as DBserver1 based on the information in the server list holding table 253. Thus, a slot group corresponding to an individual name such as DBserver1 is generated. Thereafter, when the number of servers increases or decreases, the slot group generalization unit 250 operates in the same manner.

  This slot group is also used when a server 300 that takes over the processing of the failed server is selected when any of the servers 300 in the load distribution system fails. For example, after the distribution unit 220 selects DBserver1 as the distribution destination of the received packet, when this DBserver1 fails, the server 300 that is the takeover destination of processing is immediately adjacent to the failed server (DBserver1 (A)) in the slot group. A server (for example, DBserver5 (E)) arranged in the (clockwise direction adjacent) is selected.

(Slot group generator)
Next, the slot group generation unit 241 will be described in detail with reference to FIG. The slot group generation unit 241 includes a parameter reception unit 2411, a maximum slot sequence generation request unit 2412, a determination unit 2413, a slot sequence extension unit 2414, a processing target set extraction unit 2415, a slot sequence storage processing unit 2416, A server selection unit 2417, a server replacement unit 2418, and a slot allocation unit 2419 are provided.

  The parameter receiving unit 2411 receives parameters related to slot group generation from the server group abstraction unit 251. For example, the parameter reception unit 2411 receives, as parameters from the server group abstraction unit 251, a server list (for example, A, B, C, D, E) constituting the distributed processing system, a basic permutation length (for example, 2), a maximum Information such as a list of server groups (for example, A, B, C, D, E) and the number of servers currently required (for example, 4).

  The maximum slot sequence generation request unit 2412 requests the maximum slot sequence generation unit 243 to generate a slot sequence (maximum slot sequence) of this maximum server group. Specifically, the maximum slot sequence generation request unit 2412 indicates a list of maximum server groups (for example, A, B, C, D, E) indicated in the parameters related to slot group generation, and a basic permutation length (for example, 2) is transmitted to the maximum slot sequence generation unit 243, and the maximum slot sequence is received from the maximum slot sequence generation unit 243 as a calculation result. The received maximum slot row is output to the slot row extension 2414.

  The determination unit 2413 determines whether or not server load / uninstall is necessary for the load distribution system, and if necessary, whether or not expansion / deinstallation is necessary. Specifically, the determination unit 2413 first refers to the number of currently required servers (the number of servers requested by the system maintainer 400 or the like) and the maximum number of servers in the load balancing system, which are indicated in the parameters related to slot group generation. If the currently required number of servers is smaller than the maximum number of servers, it is determined that the reduction is necessary, and if the number is the same as the maximum number of servers, it is determined that the reduction is not necessary. Further, the determination unit 2413 compares the requested number of servers with the number of servers currently in operation after starting the operation of the load balancing system, and when the requested number of servers is smaller than the number of servers currently in operation. If it is determined that a reduction is necessary and the number of requested servers is greater than the number of servers currently in operation, it is determined that an expansion is necessary.

  The slot row extension 2414 generates an extension slot row (see FIG. 13 described later) in which the maximum slot row is repeated one or more times.

  The slot row storage processing unit 2416 outputs the extended slot row generated by the slot row extension unit 2414 and the slot row generated by the server replacement unit 2418 to the slot storage unit 242.

  The processing target set extraction unit 2415 extracts a difference server group between the largest server group and the currently required server group as a server group to be removed.

  The server selection unit 2417 selects the server 300 to be replaced by the server replacement unit 2418 from the server group to be removed.

  The server replacement unit 2418 determines a location where the server 300 selected by the server selection unit 2417 appears in another slot in the extended slot sequence generated by the slot sequence extension unit 2414 or the slot sequence stored in the slot storage unit 242. Replace with 300 (perform replacement). For example, when the determination unit 2413 determines that server reduction is necessary from the largest server group of the load distribution system, the server replacement unit 2418 determines the appearance portion of the server 300 that is the target of replacement processing in the extension slot row. Replace with another server 300 appearing in the extension slot row. Details of the server replacement unit 2418 will be described later with specific examples.

  The slot allocation unit 2419 generates a slot group based on the slot sequence. For example, the slot allocating unit 2419 uses the slot space generated by the server replacing unit 2418 or the extended slot sequence generated by the slot sequence extending unit 2414 as a hash space used for distributing received packets. Divide into length slots. Then, the slot assignment unit 2419 generates a slot group in which the servers 300 are assigned to the divided slots in the order shown in the slot row (see FIG. 15). The generated slot group is output to the slot storage unit 242. When the slot sequence corresponding to the number of servers requested by the system maintainer 400 or the like is already stored in the slot storage unit 242, the slot allocation unit 2419 uses the slot sequence stored in the slot storage unit 242. Slot group.

(Maximum slot sequence generator)
The maximum slot sequence generation unit 243 will be described in detail. As illustrated in FIG. 3, the maximum slot sequence generation unit 243 includes a slot sequence permutation combination generation unit 2431, a connection graph generation unit 244, a permutation cycle extraction unit 245, and a slot sequence generation unit 246. The extended basic permutation generation unit 248 and the expansion slot sequence generation unit 249 may or may not be equipped with the maximum slot sequence generation unit 243. The case of being equipped will be described in the second embodiment. Details of the processing of each unit will be described later with specific examples.

  The slot column permutation combination generation unit 2431 generates a permutation combination of servers of (the number of assumed failure servers + 1) from the largest server group of the load distribution system.

  The connection graph generation unit 244 generates a permutation connection graph in which the permutation combinations generated by the slot column permutation combination generation unit 2431 are connected as nodes.

  The permutation cycle extraction unit 245 extracts a permutation cycle that passes through each node the same number of times from the permutation connection graph generated by the connection graph generation unit 244.

  The slot sequence generation unit 246 generates a slot sequence (basic slot sequence) by arranging the leading servers of the nodes constituting the permutation cycle extracted by the permutation cycle extraction unit 245 in the order of the cycles.

(Processing procedure)
Next, a processing procedure of the slot group management unit 240 of the load distribution apparatus 200 will be described.

(Processing procedure of maximum slot sequence generator)
First, the processing procedure of the maximum slot sequence generation unit 243 of the slot group management unit 240 will be described. As shown in FIG. 4, the maximum slot sequence generation unit 243 obtains a server list indicating servers for which a slot group is to be generated (S10), and generates a basic slot sequence for the servers indicated in the server list (S20). ).

  For example, consider a case where the maximum server group is A, B, and C, and the maximum slot sequence generation unit 243 generates a slot sequence ABCBAC by the basic slot sequence generation processing. In this case, the slot group generation unit 241 generates a slot group based on this ABCBAC slot string. For example, as shown in FIG. 10 to be described later, the slot group generation unit 241 divides the hash space into six equal parts (equally divided by the length of the slot string), and assigns servers to the slots in the same order as the slot string. That is, the slot group generation unit 241 generates a slot group of A → B → C → B → A → C (→ returns to A). In this way, the slot group generation unit 241 generates a slot group in which the number of appearances of each server in the maximum server group is the same, and the number of appearances of other servers after each server is uniform. Since the dividing unit 220 distributes the received packets according to the slot group, it is possible to equalize the load among the servers and to equalize the load at the time of failure.

  The basic slot string generation process (S20) in FIG. 4 is performed according to the procedure shown in FIG. First, the slot sequence permutation combination generation unit 2431 of the maximum slot sequence generation unit 243 generates a basic permutation based on the server list acquired in S10 of FIG. 4 (S21). Specifically, the slot sequence permutation combination generation unit 2431 extracts all permutation combinations of servers having the number of basic permutation lengths for the server group indicated in the server list (the number of assumed failure servers + 1).

For example, let us consider a case where the assumed number of failed servers is “1” and A, B, and C are described in the server list. In this case, as shown in FIG. 6, the slot sequence generation unit 243 enumerates all permutations having a length of “2 (basic permutation length)” for the three servers, and AB, AC, BC, BA, CA, CB It becomes. In other words, for example, when A fails, B takes over processing, when A fails, C takes over, when B fails, C takes over processing, and so on. List all permutations with servers that take over The permutations enumerated in this way are defined as basic permutations. When there are three servers, the number of basic permutations is ₃ P ₂ , that is, six.

  Returning to the description of FIG. After S21, the connection graph generation unit 244 generates a basic permutation connection graph (S22). Specifically, the connection graph generation unit 244 generates a basic permutation connection graph in which the permutation combinations (basic permutations) extracted in S21 are connected as nodes.

  For example, as illustrated in FIG. 7, when the basic permutation is AB, AC, BC, BA, CA, CB, the connection graph generation unit 244 has a length of 2-1 = 1 for each basic permutation. Connect to a basic permutation starting with a permutation. That is, the connection graph generation unit 244 generates a node having each basic permutation as a label, and extracts a partial permutation excluding the first one server for each basic permutation. For example, a node having a basic permutation of AB, AC, BC, BA, CA, CB as a label is generated, a partial permutation of B is extracted for AB, and a partial permutation of C is extracted for AC. Such processing is performed for all basic permutations. Then, the connection graph generation unit 244 extracts a basic permutation starting from the extracted partial permutation. For example, BC and BA are extracted as basic permutations starting from a partial permutation called B, and CA and CB are extracted as basic permutations starting from a partial permutation called C (see “Link of Basic Permutation Connection Graph” in FIG. 7). As described above, the connection graph generation unit 244 extracts a basic permutation having the same server as the end of the basic permutation as the head of the graph as a link destination of the graph. Then, the connection graph generation unit 244 connects the nodes indicating the basic permutations with the link destination nodes extracted by the above-described processing by the directed links, and generates a basic permutation connection graph as illustrated on the right side of FIG. .

  Returning to the description of FIG. After S22, the permutation cycle extraction unit 245 extracts a basic permutation cycle that passes through each node the same number of times from the basic permutation connection graph generated in S22 (S23).

  For example, the permutation cycle extraction unit 245 extracts a cycle that passes through each node the same number of times from a basic permutation connection graph as shown in FIG. 8, and obtains a basic permutation cycle shown on the right side of FIG. If such a cycle is generated, the maximum slot sequence generation unit 243 determines the order of the servers on the slot sequence along the cycle, thereby generating a slot sequence in which each basic permutation appears the same number of times. Can do. There is no limitation on the method of extracting the cycle, but at least the graph satisfies the condition having the Euler cycle, and therefore the Euler cycle can be extracted by using Fleury's algorithm.

  In the basic permutation connection graph illustrated in FIG. 8, each node is a second-order directed graph for both the incoming order and the outgoing order. Therefore, the extracted Euler circuit is a circuit that passes through each node exactly twice. In the following description, “AB → BC → CB → BA → AC → CA → AC → CB → BC → CA → AB → BA (return to AB)” shown in FIG. 8 is used as the basic permutation circuit.

  Returning to the description of FIG. After S23, the slot sequence generation unit 246 generates a slot sequence by arranging the leading servers of the nodes constituting the basic permutation cycle extracted in S23 in the order of the cycle (S24). For example, as shown in FIG. 9, the slot sequence generation unit 246 scans the basic permutation cycle in the order of AB → BC → CB → BA → AC → CA → AC → CB → BC → CA → AB → BA. The first servers are arranged along a closed path, and ABCBACABCCAB is generated as a slot train.

  Thereafter, for example, the slot group generation unit 241 equally divides the hash space by the length of the slot string generated in S24 of FIG. 5 (divides the space into 12 equal parts), and each slot in the same order as the slot string (ABCBACABCBCAB) A slot group in which servers are assigned to is generated. That is, the slot group generation unit 241 generates a slot group of A → B → C → B → A → C → A → C → B → C → A → B (→ returns to A).

  In the slot group generated in this way, all the basic permutations described above are included the same number of times. Therefore, by using such a slot group for distribution of received packets, the distribution unit 220 can equalize the load among the servers including when the server 300 in the load distribution system fails. That is, since the same number of servers 300 and permutations of the servers 300 appear in the slot group described above, the number of slots allocated in normal times is uniform. Further, for example, when there is a part assigned in the order of A → B as slot assignment, if A follows a failure according to the consistent hash method, the next B is assigned. As a result, when the number of appearances of all the basic permutations is the same, the number of times other servers 300 appear after each server 300 becomes uniform, and the load can be made uniform even at the time of failure.

  In the above, for the sake of simplicity, the procedure for generating the slot train when the maximum number of servers in the load balancing system is three, A, B, and C, is shown, but the maximum number of servers is A, B, The slot sequence for C, D, and E is generated in the same procedure. For example, if the maximum number of servers is 5 (A, B, C, D, E), the maximum slot sequence generator 243 generates “ABCDEABDADCDBDBEDEDCACBAEBDDECABEACDCBEBCCEBACEADEBEDECECBCBCBDACADAEDEDEDAECADB in each node instead of in a closed node” Is used, a slot sequence "AEADEDAEABEBDEDBEBCDECDCCBDBADACDCACBCAB" is generated. Then, the maximum slot sequence generation unit 243 outputs the generated slot sequence to the slot group generation unit 241. In the following explanation, the latter “AEADEDAEABEBDEDBEBBCDECDCDBADACDCACBCAB” is used as the maximum slot row.

(Processing procedure of slot group generation unit)
A processing procedure of the slot group generation unit 241 will be described with reference to FIG. When the parameter reception unit 2411 of the slot group generation unit 241 receives the parameter related to slot group generation from the server group abstraction unit 251 (S1), the maximum slot sequence generation request unit 2412 sends the maximum slot sequence generation unit 243 to the maximum slot. A column generation request is made (S2). Thereafter, when the maximum slot sequence is received from the maximum slot sequence generation unit 243 (S3), the determination unit 2413 indicates the maximum number of server groups (maximum number of servers) and the currently required server indicated by the parameter received in S1. The number is compared to determine whether or not the server 300 needs to be removed (S4). If the determination unit 2413 determines that the server 300 needs to be removed (Yes in S4), the slot group generation unit 241 generates a reduction slot group (S5). That is, the slot group generation unit 241 generates a reduced slot row, and generates a slot group (removed slot group) based on the generated reduced slot row. On the other hand, if the determination unit 2413 determines that the server 300 need not be removed (No in S4), the slot group generation unit 241 generates a slot group based on the maximum slot sequence received in S3 (S6). .

  The reduced slot group generation process (S5) in FIG. 11 is performed according to the procedure shown in FIG. First, the slot row extension unit 2414 generates an extension slot row that is repeated one or more times for the maximum slot row obtained from the maximum slot row generation unit 243 (S80: slot row extension). For example, the slot row extension 2414 generates an extension slot row that is repeated three times as shown in FIG. The number of repetitions may be 1, but basically, the larger the number of repetitions, the more the load unevenness among the servers 300 of the load distribution system can be suppressed.

  Next, the processing target set extraction unit 2415 reduces the difference between the server set appearing in the extended slot sequence generated in S80 of FIG. 12 and the server group that is requested to generate the slot group, and the server 300 that is to be reduced. (S90: Extraction target set extraction). For example, the processing target set extraction unit 2415 extracts {E} as a reduction target set from the largest server group (A, B, C, D, E), for example. After S90, the server selection unit 2417 selects one server at a time until the reduction target set becomes an empty set.

  That is, if there is a reduction target in the reduction target set (Yes in S100), the slot row storage processing unit 2416 first stores the extended slot row at that stage in the slot storage unit 242 (S110: slot row storage). . Here, the extension slot row is stored because when the server 300 is added after the server 300 is removed and the operation of the load balancing system is started, a slot group including the added server 300 is generated. This is for occasional use. On the other hand, if there is no reduction target in the reduction target set in S100 (No in S100), the slot group generation unit 241 generates a slot group based on the current slot sequence (S150). Then, the generated slot group is output to the slot storage unit 242.

  After S110, the server selection unit 2417 selects one target server 300 for replacement processing from the reduction target set {E} under conditions such as descending order (S120: processing target server selection). For example, the server selection unit 2417 selects E from the reduction target set {E}.

  After S120, the server replacement unit 2418 replaces the server 300 (for example, E) to be replaced in the extension slot row with another server 300 (S130: processing target server replacement). Specifically, the server replacement unit 2418 identifies the location where the server 300 selected in S120 (for example, E) appears in the extension slot row as another server 300 (for example, A, B, C, etc.) constituting the load balancing system. , D). At this time, the server replacement unit 2418 makes the number of appearances of each of the basic permutations (for example, AB, CD, etc.) of the servers 300 (A, B, C, D) appearing in the extension slot row after server replacement as much as possible. Replace as follows. Details of the processing of S130 will be described later with specific examples.

  After S130, the server replacement unit 2418 updates the reduction target set (S140), and returns to S100. If there is no reduction target in the reduction target set in S100 (No in S100), the slot group generation unit 241 generates a slot group based on the current slot sequence (S150). That is, the slot group generation unit 241 generates a slot group by dividing the hash space by the length of the generated slot string and assigning it to each server 300 along the slot string. For example, when the length of the generated slot sequence is “20”, the hash space is divided into 20 as illustrated in FIG. 15 and assigned to each server 300 along the generated slot sequence. Create a group.

  In this slot group, the load when each server 300 is operating normally becomes more uniform as the number of appearances of each server 300 is the same among the servers 300. Further, the more the same number of appearances of each basic permutation is in this slot group, the more uniform the load among the active servers when the basic permutation length is reduced to one.

(Processing server replacement process)
The processing target server replacement process in S130 of FIG. 12 will be described in detail. For example, let us consider a case where E is selected as a replacement processing target from A to E in the extension slot row as shown in FIG. In this case, E appears 24 times in the extended slot row, but the server replacement unit 2418 selects one of the servers other than E (any one of A, B, C, and D) at the appearing location. Replace with. At this time, the server replacement unit 2418 performs replacement so that the number of appearances of each basic permutation (for example, AB, CD, etc.) is the same as much as possible.

  For example, the server replacement unit 2418 replaces each of the locations where E appears in the extension slot sequence with A, B, C, and D, and after replacement, all the basic permutations (AB, BA, BC, CB, (CD, DC, DA, AD, AC, CA, BD, DB) is searched for a replacement with the smallest variance value of the number of appearances.

  As a result, for example, as shown in FIG. 14, from the extended slot row with 5 servers, a slot row with 4 servers shown in FIG. 14 is generated. In this slot row with 4 servers, all the basic permutations (AB, BA, BC, CB, CD, DC, DA, AD, AC, CA, BD, DB) appear equally 10 times.

  Therefore, if the slot allocation unit 2419 generates a slot group along this slot row, any one of the four servers A, B, C, and D operates normally in the load balancing system. It is possible to generate a slot group that equalizes the load between servers even when a failure occurs. In addition, the server replacement unit 2418 generates a slot train (removal slot train) in which the location where the server 300 to be removed appears is replaced with another server 300 from the extended slot train. Therefore, if the slot allocation unit 2419 generates a slot group along this slot row, even if the server 300 is reduced, the hash of the server 300 to be reduced is compared with the slot group before the reduction. A slot group in which only the space allocation range is changed can be generated. For example, even when the number of servers constituting the load balancing system fluctuates between 4 (A, B, C, D) and 5 (A, B, C, D, E), the server 300 ( For example, it is possible to generate a slot group in which only the assigned range of the hash space assigned to E) is changed. Therefore, even when the number of servers constituting the load balancing system increases or decreases, the distribution unit 220 distributes packets using the slot group generated by the above processing, so that the stored data between the servers It is possible to minimize the influence due to the movement of the.

  After this, if there is a request from the system maintenance person 400 or the like regarding the change of the server 300 that constitutes the load balancing system, the slot group generation unit 241 that has received the request is requested the number of servers in operation. It is determined whether or not the number of servers 300 needs to be reduced or increased.

  For example, when the number of servers currently in operation is 4 (A, B, C, D) and the requested number of servers is 3 (A, B, C), the slot group generation unit 241 that has received the request. Determines that the server 300 (for example, D) needs to be removed. Then, the slot group generation unit 241 does not execute the slot string extension process (S80) in the flow shown in FIG. 12, but executes the processes after S90, thereby the number of servers stored in the slot storage unit 242. From the slot row in the case of 4, a slot row in which the reduction target server 300 (for example, D) is removed is generated. For example, the slot group generation unit 241 generates a slot string in the case of 3 servers shown in FIG. And the slot group production | generation part 241 produces | generates the slot row | line | column which reduced the server 300 (for example, D) of reduction object.

  Further, for example, when the number of servers currently in operation is 3 and the number of servers requested is 4 or 5, that is, the number of servers requested is larger than the number of servers currently in operation, and When the number is smaller than the maximum number of servers, the slot group generation unit 241 that has received the request determines that the server 300 needs to be added. Then, the slot group generation unit 241 takes out the slot train in the case of the number of servers 4 or 5 (for example, the slot row of the number of servers 4 or the slot row of the number of servers 5 in FIG. 14) from the slot storage unit 242, and Based on the above, a slot group is generated.

  By doing so, for example, even when the number of servers in the load balancing system is increased from 3 (for example, A, B, C) to 4 (for example, A, B, C, D), the slot group generation unit 241 can generate a slot group that does not change the assigned range of the hash space assigned to servers 300 (for example, A, B, C) other than the server 300 (for example, D) to be added as much as possible. As a result, even when the number of servers 300 is increased in the load balancing system, if the number of servers is less than or equal to the maximum number, the slot group generation unit 241 has an effect due to the movement of stored data between servers while equalizing the load among servers. Slot groups that can be minimized are generated.

(Second Embodiment)
Next, a second embodiment of the present invention will be described. In the second embodiment, when the maximum slot sequence generation unit 243a generates a slot sequence of the maximum number of slots (maximum slot sequence), it generates slot sequences of some servers in the server list (basic slot sequence generation). The slot sequence is expanded to the remaining servers in the server list (slot sequence expansion is performed). Thus, even when the number of servers shown in the server list is large, the maximum slot sequence generation unit 243a generates a slot sequence by combining basic slot sequence generation and slot sequence expansion. The amount of calculation and the slot length can be reduced.

  Hereinafter, the maximum slot sequence generation unit 243a first generates a slot sequence of (basic permutation length + 1) servers among the servers in the server list, and then extends this slot sequence to the remaining servers. Will be described as an example. The same configurations as those of the first embodiment described above are denoted by the same reference numerals and description thereof is omitted.

  As shown in FIG. 3, the maximum slot sequence generation unit 243a further includes an extended basic permutation generation unit 248 indicated by a broken line and an expansion slot sequence generation unit 249.

  First, the maximum slot sequence generation unit 243a includes a server (for example, A to E) indicated in the server list by the slot sequence combination generation unit 243, the connection graph generation unit 244, the permutation cycle extraction unit 245, and the slot sequence generation unit 246. ), A slot sequence of (basic permutation length + 1) servers, for example, three servers A, B, and C is generated.

  The extended basic permutation generation unit 248 is a permutation combination of (basic permutation length-1) servers from the servers (for example, A, B, and C) indicated in the basic permutation generated by the slot permutation combination generation unit 243. Extract all extended basic permutations. For example, the extended basic permutation generation unit 248 extracts A, B, and C as the extended basic permutation.

  The expansion slot string generation unit 249 selects one of the servers (for example, A to E) indicated in the server list, which is not targeted by the slot string generation unit 246 as a slot string (basic slot string).

  For example, the expansion slot sequence generation unit 249 selects either D or E, and the expansion slot sequence generation unit 249 uses the expansion basis from the slot sequence (for example, ABCBAC) generated by the slot sequence generation unit 246. One server (for example, A, B, C) shown in the permutation is extracted one by one, and each expansion target server (for example, D) is replaced with a permutation sandwiched between expansion basic permutations (for example, A, B, C). . As a result, a new slot sequence in which the expansion target server is inserted is generated in the basic slot sequence. That is, the basic slot sequence is expanded to the expansion target server. Thereafter, the expansion slot string generation unit 249 inserts an expansion target server (for example, E) further into the slot string in which the expansion server (for example, D) is inserted, by the same processing as described above. As a result, a slot train of all servers (A, B, C, D, E) shown in the server list is generated.

  Details of the extended basic permutation generation unit 248 and the expansion slot sequence generation unit 249 will be described later using a specific example.

  Next, a slot group generation processing procedure in the maximum slot string generation unit 243a of the second embodiment will be described.

  As shown in FIG. 16, the maximum slot sequence generation unit 243a acquires basic information indicating the server list and the basic permutation length (S40), and extracts the server list for generating the basic slot sequence based on the basic information. (S50). Then, based on the server list extracted in S50, the maximum slot string generation unit 243a generates a basic slot string in the same procedure as S20 in FIG. 4 (S20). Then, the maximum slot sequence generation unit 243a generates extension information (S60). This extension information is information used to extend the slot sequence, and includes a server list for generating a basic slot sequence (pre-expansion server list), a basic slot sequence (slot sequence before expansion) generated in S20, This is information indicating the basic permutation length and a list of servers (extended server list) excluded from the basic slot sequence in the server list.

  For example, as illustrated in FIG. 17, when the server list indicated in the basic information is “A, B, C, D, E” and the basic permutation length is “2”, the maximum slot sequence generation unit 243 a Basic permutation length + 1 = 3 servers are arbitrarily extracted, and a server list (for example, A, B, C) for generating a basic slot sequence is extracted. Here, servers that are not subject to extraction (for example, D and E) are put on the extended server list. Thereafter, the maximum slot string generation unit 243a generates a slot string by a basic slot string generation process (a process similar to S20 in FIG. 4). This slot row is a basic slot row. The maximum slot sequence generation unit 243a has the pre-expansion server list (server list for basic slot sequence generation) “A, B, C”, and the slot sequence before expansion (basic slot sequence) is “ABCBAACACBCAB”. Yes, extended information indicating that the basic permutation length is “2” and the extended server list is “D, E” is generated.

  Returning to the description of FIG. After S60, the maximum slot string generation unit 243a executes a slot string expansion process based on the expansion information generated in S60 (S70).

  Here, the slot string expansion process in S70 of FIG. 16 will be described with reference to FIG. When the extended basic permutation generation unit 248 acquires the extended information generated in S60 (S71), the extended basic permutation generation unit 248 extracts an expansion target server (S72) and generates an extended basic permutation (S73).

For example, the extended basic permutation generation unit 248 extracts D from the extended server list “D, E” of the extended information shown in FIG. Then, as shown in FIG. 19, the extended basic permutation generation unit 248 uses three servers A, B, and C from the information that the pre-extension server list is “A, B, C” and the basic permutation length is “2”. For all permutations with a basic permutation length 2-1 = 1. The permutation extracted here is called an extended basic permutation. For example, the extended basic permutation generation unit 248 generates “A, B, C” as the extended basic permutation. The number of extended basic permutations extracted here is ₃ P ₁ = 3 when there are three servers A, B, and C.

  Returning to the description of FIG. After S73, the expansion slot string generation unit 249 inserts the expansion basic permutation generated in S73 into the slot string (S74).

For example, the expansion slot string generation unit 249 searches the position of the expansion basic permutation (for example, A, B, C) generated in S73 two times from the slot string before expansion (for example, ABCBACABCBCAB). Go. The number of times of searching is determined by dividing the slot sequence length before expansion by the number of basic permutations, which is equal to the number of passes through each node in the basic permutation cycle of the slot sequence (see FIG. 8). . For example, the above-mentioned means if the length of the extended previous slot columns (ABCBACACBCAB) in Example "12", the number of basic permutations since the _"3 P ₂ = 6", and "12 ÷ 6 = 2".

  Then, the expansion slot string generation unit 249 replaces each expansion basic permutation searched from the slot string before expansion with a permutation sandwiching the expansion target server in the expansion basic permutation. For example, consider a case where the expansion target server is “D”. In this case, as illustrated in FIG. 20, the expansion slot sequence generation unit 249 replaces the expansion basic permutation “A” with “ADA” for two locations that are “A” in the slot sequence before expansion. To do. Similarly, two locations that are “B” in the slot row before expansion are replaced with “BDB”. Furthermore, “CDC” is replaced with two locations that are “C” in the slot row before expansion. As a result, “ADABDBCDCBDBADACDCACBCAB” is obtained as the expanded slot string.

  In this way, the expansion slot string generation unit 249 extracts two expansion basic permutations from the slot string before expansion, and replaces each expansion target server with a permutation sandwiched between the expansion basic permutations. In the above example, the case where the length of the extended basic permutation is “1” has been described. However, when the length is “2”, for example, when the extended basic permutation is “AB”, the extension target If the server is “D”, the place “AB” in the slot string before expansion is replaced with “ABDAB”.

  Returning to the description of FIG. After inserting the expansion basic permutation into the slot string in this way (S74), the expansion slot string generating unit 249 updates the extension information by adding the processed expansion target server to the server list before expansion (S75). . For example, the expansion slot string generation unit 249 inserts the server “D” into the slot string “ABCBACABCBCAB” before expansion, and then updates the left extension information in FIG. 21 to the right extension information.

  Returning to the description of FIG. If the expansion target server remains in S76 (Yes in S76), the maximum slot string generation unit 243a executes the processing from S71 again using the updated expansion information. On the other hand, if the expansion target server does not remain (No in S76), the slot string expansion process is terminated.

  The expansion slot string generation unit 249 inserts the expansion target server “D” into the slot string by the process shown in FIG. 20, and then inserts the expansion target server “E” into the slot string by the same process as described above. That is, the expansion slot string generation unit 249 generates a slot string “ADABDBCDCBDBADACDCACBCAB”, then extracts two locations where “A, B, C, D” of this slot string appears, and expands each of them. Replacement is performed with a permutation sandwiching the server “E”. As a result, the expansion slot sequence generation unit 249 generates a slot sequence of “AEADEDAEABEBDEDBEBCDECDECBDBADACDCACBCAB”.

  When the slot sequence incorporating all the servers (A, B, C, D, E) shown in the server list is completed by the above processing, the slot group generation unit 241 generates a slot group based on this.

  For example, when the maximum slot string generation unit 243a generates a slot in which the space of the hash value is divided into the length 40 of the slot string “AEADEDAEABEBDDECBBECDCDECDBADACDCACBCAB”, the slot group generation unit 241 includes the slot string “AEADEDAEABEBDBCBDECDCACDCCBDCDC” in this slot. Assign the servers in order.

  As a result of the above processing, a slot group including all the servers shown in the server list and having a uniform assigned range among them is generated. Therefore, in the load balancing system using the slot group, the load is distributed between the servers. It becomes possible to disperse uniformly.

  In addition, the maximum slot sequence generation unit 243a limits the generation of slot sequences to (basic permutation length + 1) units, and inserts servers that exceed (basic permutation length + 1) units into the slot sequence by slot sequence expansion processing. Thus, it is possible to reduce the amount of calculation and the slot length when generating the slot train for all the servers shown in the server list.

(Other embodiments)
The slot group management unit 240 in FIG. 1 is illustrated as being one in the load distribution apparatus 200, but is cut out from the load distribution apparatus 200 as shown in FIG. In this case, one slot group providing apparatus 500 can perform processing of a plurality of load distribution apparatuses 200.

(program)
It is also possible to create a program in which the processing executed by the load distribution apparatus 200 according to the above embodiment is described in a language that can be executed by a computer. In this case, the same effect as the above-described embodiment can be obtained by the computer executing the program. Further, such a program may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read by a computer and executed to execute the same processing as in the above embodiment. Hereinafter, an example of a computer that executes a load distribution program that implements the same unit as the load distribution apparatus 200 will be described.

  FIG. 23 is a diagram illustrating a computer that executes a load distribution program. As shown in FIG. 23, the computer 1000 includes, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These units are connected by a bus 1080.

  The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM (Random Access Memory) 1012. The ROM 1011 stores a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. A removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1100, for example. For example, a mouse 1110 and a keyboard 1120 are connected to the serial port interface 1050. For example, a display 1130 is connected to the video adapter 1060.

  Here, as shown in FIG. 23, the hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. Each table described in the above embodiment is stored in the hard disk drive 1090 or the memory 1010, for example.

  Further, the load distribution program is stored in the hard disk drive 1090 as a program module in which a command executed by the computer 1000 is described, for example. Specifically, a program module describing each process executed by the load distribution apparatus 200 described in the above embodiment is stored in the hard disk drive 1090.

  Further, data used for information processing by the load balancing program is stored as program data in, for example, the hard disk drive 1090. Then, the CPU 1020 reads out the program module 1093 and the program data 1094 stored in the hard disk drive 1090 to the RAM 1012 as necessary, and executes the above-described procedures.

  Note that the program module 1093 and the program data 1094 related to the load balancing program are not limited to being stored in the hard disk drive 1090. For example, the program module 1093 and the program data 1094 are stored in a removable storage medium and read by the CPU 1020 via the disk drive 1100 or the like. May be issued. Alternatively, the program module 1093 and the program data 1094 related to the load distribution program are stored in another computer connected via a network such as a LAN (Local Area Network) or a WAN (Wide Area Network), and are transmitted via the network interface 1070. May be read by the CPU 1020.

DESCRIPTION OF SYMBOLS 100 Client 200 Load distribution apparatus 210 Packet receiving part 220 Distribution part 221 Key extraction part 222 Packet transmission destination determination part 230 Packet transfer part 240 Slot group management part 241 Slot group generation part 242 Slot storage part 243, 243a Maximum slot row generation part 244 connection graph generation unit 245 permutation cycle extraction unit 246 slot sequence generation unit 248 extended basic permutation generation unit 249 expansion slot sequence generation unit 250 slot group generalization unit 251 server group abstraction unit 252 server group individualization unit 253 server list holding table 300 server 2411 parameter reception unit 2412 maximum slot sequence generation request unit 2413 determination unit 2414 slot sequence extension unit 2415 processing target set extraction unit 2416 slot sequence storage processing unit 2417 server selection unit 241 Server replacement section 2419 slot assignment unit 2431 slot bank permutations generator

Claims (8)

In a load distribution system that distributes load among a plurality of servers, a slot group generation device that generates a slot group used for distributing received packets,
Maximum slot sequence generation for generating a slot sequence in which servers are arranged so that the number of appearances of permutation combinations of a predetermined number of servers is the same for each maximum server group when the number of servers constituting the load balancing system is maximized And
An extension slot string generation unit that generates an extension slot string obtained by duplicating and connecting one or more of the generated slot strings;
When the number of servers required for the load balancing system is smaller than the number of servers in the maximum server group, the appearance of servers to be removed from the maximum server group in the generated extension slot row A server replacement unit that generates a slot sequence in which each portion is replaced with a server that is not subject to reduction so that the number of appearances of each basic permutation of the server that is not subject to reduction is the same as possible,
A predetermined hash space is divided into slots each having the length of the slot sequence generated by the server replacement unit, and a slot group in which the servers are assigned to each of the divided slots in the order indicated by the slot sequence is generated. A slot group generation device comprising a slot allocation unit. The slot group generation device further includes:
A storage unit for storing the extended slot row and the slot row generated in the process of replacement by the server replacement unit;
The slot allocation unit includes:
When there is a slot sequence corresponding to the number of servers required for the load distribution system in the storage unit, the slot sequence corresponding to the number of servers required for the load distribution system is obtained from the storage unit, and the slot The slot group generation device according to claim 1, wherein a group is generated. The server replacement unit
There is no slot sequence corresponding to the number of servers required for the load distribution system in the storage unit, and the number of servers required for the load distribution system is smaller than the number of servers in the maximum server group. When the slot string corresponding to the current server group of the load distribution system is acquired from the storage unit, the appearance of the server to be removed from the server group of the current load distribution system in the acquired slot string 3. The slot train in which each portion is replaced with a server that is not subject to reduction is generated so that the number of appearances of each basic permutation of the server that is not subject to reduction is the same as possible. Slot group generator. The maximum slot sequence generation unit includes:
A slot permutation combination generation unit that generates a basic permutation that is a permutation combination of a predetermined number of servers from the maximum server group when the number of servers constituting the load balancing system is maximized;
A connection graph generation unit for generating a permutation connection graph in which the generated basic permutation is connected as a node;
From the generated permutation connection graph, a permutation cycle extraction unit that extracts a permutation cycle in which the number of times each node passes is the same between all nodes;
4. A slot sequence generation unit configured to generate a slot sequence of the largest server group by arranging the first servers of the nodes constituting the extracted permutation cycle in the order of the cycle. 2. A slot group generation device according to item 1. The predetermined number is
The slot group generation device according to claim 1, wherein the slot group generation device is a value obtained by adding 1 to the number of failed servers assumed in the load distribution system. The slot group generation device includes:
The maximum slot server sequence generator is excluded from the slot bank, when incorporated into slot column the maximum slot sequence generating unit has generated, the server indicated in the slot column the maximum slot sequence generating unit has generated, the An extended basic permutation generating unit that generates an extended basic permutation that is a permutation combination of a number of servers obtained by subtracting 1 from a predetermined number;
A permutation in which the servers excluded from the slot sequence are included in the extended basic permutation is generated, and the location of the extended basic permutation in the slot sequence generated by the maximum slot sequence generation unit is replaced with the generated permutation. The slot group generation device according to claim 1, further comprising: an expansion slot sequence generation unit that generates a slot sequence. In a load distribution system that distributes load among a plurality of servers, a slot group generation device that generates a slot group used for distribution of received packets includes:
Maximum slot sequence generation for generating a slot sequence in which servers are arranged so that the number of appearances of permutation combinations of a predetermined number of servers is the same for each maximum server group when the number of servers constituting the load balancing system is maximized Steps,
An extended slot sequence generating step of generating an extended slot sequence in which one or more of the generated slot sequences are duplicated and connected;
When the number of servers required for the load balancing system is smaller than the number of servers in the maximum server group, the appearance of servers to be removed from the maximum server group in the generated extension slot row A server replacement step for generating a slot train in which each portion is replaced with a server that is not subject to reduction so that the number of appearances of each basic permutation of the server that is not subject to reduction is the same as possible,
A predetermined hash space is divided into slots each having a length of the slot string generated by the server replacement step, and a slot group in which the servers are assigned to the divided slots in the order indicated by the slot string is generated. A slot group generation method comprising: performing a slot allocation step. In a load distribution system that distributes load among a plurality of servers, a slot group generation device that generates a slot group used for distribution of received packets, and distribution that distributes the received packets using the generated slot groups A load balancing system comprising a device,
The slot group generation device includes:
Maximum slot sequence generation for generating a slot sequence in which servers are arranged so that the number of appearances of permutation combinations of a predetermined number of servers is the same for each maximum server group when the number of servers constituting the load balancing system is maximized And
An extension slot string generation unit that generates an extension slot string obtained by duplicating and connecting one or more of the generated slot strings;
When the number of servers required for the load balancing system is smaller than the number of servers in the maximum server group, the appearance of servers to be removed from the maximum server group in the generated extension slot row A server replacement unit that generates a slot sequence in which each portion is replaced with a server that is not subject to reduction so that the number of appearances of each basic permutation of the server that is not subject to reduction is the same as possible,
A predetermined hash space is divided into slots each having the length of the slot sequence generated by the server replacement unit, and a slot group in which the servers are assigned to each of the divided slots in the order indicated by the slot sequence is generated. A load distribution system comprising a slot allocation unit.

Images