JP3784054B2 - Method and recording medium for rearranging MAC addresses - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a MAC address pointer structure and a MAC address rearrangement method for quickly searching an arbitrary MAC address from a large number of MAC (Media Access Control) addresses in a large-scale switch or the like used in a network. .
[0002]
[Prior art]
A system that can support hundreds of thousands of MAC addresses in an L2 switch used in a conventional network can support many servers and clients, but retains it to identify the corresponding port from the MAC address. After searching all the MAC addresses, a port corresponding to the MAC address must be found. This is a major factor in reducing performance for a large-scale switch having a MAC address exceeding hundreds of thousands.
[0003]
In order to search an arbitrary MAC address from a MAC address table exceeding hundreds of thousands at high speed, there are the following conventional methods.
(1) Having a cache memory that can temporarily hold a MAC address, and when the received destination MAC address exists in the cache memory, a port is derived from the cache memory, and transfer is performed only to that port.
(2) However, when the cache memory does not hold the destination MAC address, it is necessary to transfer to all ports except the transmission port by broadcast.
[0004]
In the case of (1), since it is impossible to store a MAC address whose cache size exceeds several hundred thousand, when a network host having a cache size or larger communicates, the frequency of performing the broadcast of (2) is high. Become. Since this broadcast performs a process of placing packets simultaneously on all the related networks, the load on the related networks increases. As a result, the network is in a state of operation like a hub that temporarily stops all the networks, resulting in a decrease in performance.
[0005]
In addition, when CAM is used as another conventional method, CAM cannot store a MAC address table exceeding hundreds of thousands, so it is necessary to perform high-speed MAC address search using a normal memory. There are various possible search methods. If hundreds of thousands of MAC addresses are arranged in a manner adapted to the search method, the search can be performed at high speed.
[0006]
However, hundreds of thousands of MAC addresses are not always arranged in an easy-to-search state. Some network hosts with hundreds of thousands of connected states may not always have the same connection location as in mobile products. Therefore, the relationship between some MAC addresses and the ports of switches and routers changes with time, and there is a possibility that data (address) is constantly added and deleted. The rearrangement of data in the MAC address table accompanying the addition and deletion of data in the MAC address table must be performed periodically. This rearranging operation requires a time of several milliseconds when the number of entries exceeds several hundred thousand, and it is not realistic to perform a MAC address search using a normal memory.
[0007]
[Problems to be solved by the invention]
An object of the present invention is to retrieve an arbitrary MAC address from a large number of MAC addresses in a large-scale switch or the like used in a network at high speed.
[0008]
An object of the present invention is to provide a MAC address pointer structure and a MAC address rearrangement method that enable a MAC address to be searched at high speed.
[0009]
[Means for Solving the Problems]
The present invention is a MAC pointer table structure having an N-bit entry number selected from among bits constituting a MAC address, each entry including a pointer for designating an MAC address entry table, the pointer being It is a MAC pointer table structure consisting of N bits of a MAC address.
[0010]
The present invention is also a method for rearranging MAC addresses,
(A) preparing a MAC address entry table;
(B) preparing a MAC pointer table;
(C) obtaining MAC address entry data;
(D) checking the MAC pointer table;
(E) storing data at the address of the entry table indicated by the pointer of the checked MAC pointer table, and reordering the MAC addresses.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
The present invention comprises three features: (1) “segmentation of entry table”, (2) “pointer table structure for directly designating a segmented table”, and (3) “exception processing”. Hereinafter, these features will be described.
[0012]
(1) “Subdivision of entry table” Before describing the subdivision of the entry table, the characteristics of the premise of the MAC address will be described.
The MAC address is 48 bits in combination of a 24-bit vendor ID and a 24-bit local assigned number. The vendor ID is currently allocated about 5000, and is not allocated continuously. On the other hand, the local assignment number cannot be generally determined whether it is continuously assigned by the manufacturing company, but one or more (usually tens of thousands or more) is assigned to one vendor ID. . Therefore, if a local assignment number can be used to indicate a subdivided entry table and all MAC addresses can be distributed and stored in the subdivided entry table, the subaddresses are stored directly from the MAC address. This means that a table can be specified.
[0013]
When a few bits from the lower bits of the local assignment number are used as a pointer to specify a subdivided table, the number of entries in each subdivided table tends to be distributed as the number of MAC addresses increases. Come. When actual data is applied, it may be concentrated on other tables than on a small number of tables, but it is distributed almost evenly on many tables. This is because the randomness of the local assignment number appears as the number of MAC addresses increases. In some tables, MAC address entry data may be concentrated. However, as long as exception processing is performed for this concentration, each table can be directly specified from the MAC address.
[0014]
For example, about 10000 MAC address data,
(A) Dispersion tendency when 12 bits from the lower bit are used as a pointer (b) Dispersion tendency when the 13th and 14th bits are used (c) Dispersion tendency of 12 bits from the third bit (d) ) Distribution tendency when 12 bits of 0, 1 ratio close to 50:50 are used as pointers for each bit (e) The distribution tendency is examined more than the distribution tendency by the hash function for the MAC address currently disclosed As a result, almost the same dispersion tendency was obtained. That is, except for some tables, it was found that the number of collisions of entry data is distributed in 0-2. This is consistent with the fact that random data can be obtained when the number of samples in statistics exceeds a certain level.
[0015]
The subdivision of the entry table derives a necessary table size from the MAC address search allowable time in order to shorten the search and rearrangement time. For example, when a search speed corresponding to Gigabit Ethernet (R) is required, if the minimum packet size is 72 bytes, if a search can be performed between receiving the destination and the source address and receiving the FCS (52 bytes), The occurrence of processing delay can be prevented by the MAC address search. Assuming that the read cycle is 20 nS, the maximum number of times that information for the destination and the source address can be obtained is (52/2) * 8/20 = 10 times. When the MAC address is entered into the table, the insertion position is determined at the time of retrieval, so the maximum movement amount in the data table is half the size of the table. Considering moving within the table by a combination of read and write, the table size of entry 8 is determined from “number of searches (within 2 times for the maximum movement amount)” plus “movement time”.
[0016]
Finally, in order to search the MAC address at high speed, the number of entries in the MAC address table is 8 or less. As a result, when searching using the two-branch search method, if a search and comparison is performed a maximum of four times, it is possible to obtain information on whether the MAC address exists and where the transfer destination port is. Since the entry table itself is reduced by subdivision and the number of entries can be reduced, the sorting of entry data can be accelerated, for example, when using a two-branch search.
[0017]
(2) A table fragmented for “pointer table structure that directly designates a fragmented table” is designated by the MAC pointer table. Figure 1 shows the basic operation using the MAC pointer table.
It is a figure explaining (configuration). The size of the MAC pointer table 10 is a size obtained by subtracting the size (for example, 3 bits) of the entry table 12 from the total number of MACs (for example, 17 bits) (for example, 14 bits). Each entry 14 has a bit array configuration 16 shown in FIG. The arrangement address for each entry 14 in the MAC pointer table 10 uses the lower 14 bits of the local assignment number 20 of the MAC address 18. This corresponds to a hash function for decentralization. The EXIST bit 22 (1 bit) indicates whether or not there is an entry of MAC address data for the lower 14 bits of the local assignment number 20. When the EXIST bit 22 is active (“1”), the memory pointer (14 bits) of the entry 22 is valid, which indicates each head address of the fragmented entry table (24 in FIG. 1).
[0018]
(3) About “exception handling”
(A) When the distribution tendency does not appear in the lower bits of the local assignment number "
(B) There are two types, “when there is entry data exceeding the table size”.
(A) Processing when the distribution tendency does not appear in the lower bits of the local assignment number When the MAC address entry data is not distributed by the lower 14 bits (the number is an example of this time), other than the lower 14 bits of the local assignment number Dispersion is performed by combination. For example,
(I) Take 14 bits from the lower 4th bit,
(Ii) For each bit, 14 bits out of bits having a ratio of 0 and 1 close to 50:50 are used as arrangement addresses for each entry.
Etc.
[0019]
In order to realize these, the pointer mask register 30 of FIG. 2 is provided. This is used to designate any 14 bits of the local assignment number. The example of FIG. 2 shows the register 30 that selects from the local assignment number 24 bits, but a structure that selects 14 bits from 48 bits of the MAC address is also possible. After that, the entry 14 of the MAC pointer table 10 can specify the segment table 12 subdivided by having the structure 16 of FIG. 2A as in the above (2) pointer table structure.
[0020]
(B) Processing when there is entry data larger than the table size Since there is no complete hash function, entry data larger than the table size may exist for some tables. In this case, the configuration shown in FIG. 3 is adopted.
(I) When entry data larger than the table size exists, the entry data of the table must be divided into two tables. As a flag used for distribution for each MAC address for this purpose, an arbitrary bit out of 10 unused bits among the 24 bits of the local assignment number is used.
(Ii) The pointer mask register 32 of FIG. 3A is the same as that of FIG. 3A (FIG. 2), and 10 bits other than the 14 bits designated by this register 32 are replaced with the 2nd pointer mask of FIG. 34. The 2nd pointer memory address 36 includes information 38 ((3) in FIG. 3) for designating an entry data table divided by the bit (address) designated by the “pointer mask register 32” plus the “2nd pointer mask 34”. This is the start address of the stored memory.
(Iii) The additional MAC pointer table entry (3) designated by the 2nd pointer memory address 36 has the same configuration as the MAC pointer table entry (see 16, (1) in FIG. 2) in the basic operation.
[0021]
FIG. 4 is a diagram showing a flow of a data entry table search operation using the above-described MAC pointer table structure of the present invention. In FIG. 4, from start 400 to reference numeral 424 shows the flow of basic operations with the configuration of FIG. 1, and reference numerals 500 to 506 show the flow of exception processing with the configuration of FIG. 2 or FIG. Also, reference numerals 412 to 420 in FIG. 4 indicate a flow for determining whether or not processing can be performed within the search allowable time. As described above, according to the present invention, each entry table is designated from the MAC address using a pointer, and the table itself is subdivided, thereby realizing high-speed search and rearrangement.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of an embodiment of a MAC pointer table structure according to the present invention.
FIG. 2 is a diagram showing a configuration of an embodiment of a MAC pointer table structure according to the present invention.
FIG. 3 is a diagram showing a configuration of an embodiment of a MAC pointer table structure according to the present invention.
FIG. 4 is a diagram showing a flow of a data entry table search operation using the MAC pointer table structure of the present invention.
[Explanation of symbols]
10: MAC pointer table 12: entry table 14: one entry 16 in the entry table 12: bit array configuration 18 of the entry 14: MAC address 20: local assignment number 22: EXTIT bit 24: address indication 30, 32: pointer mask register 34: 2nd pointer mask 36: 2nd pointer memory address 38: 1 entry of additional pointer table

Claims (11)

A method for causing a computer to rearrange MAC addresses,
(A) a step of Ru to produce an entry table of the MAC address having a predetermined number of entries that are determined from the table size derived from the search time allowed MAC addresses,
(B) a step of Ru to produce a MAC pointer table having a number of entries in the N bits in the bits constituting the MAC address,
(C) a step of Ru to acquire entry data of the MAC address,
(D) determining from the bit information of each entry of the MAC pointer table whether the pointer of each entry is valid or invalid ;
(E) a step of said determined pointer to be effective for MAC pointer table Ru is stored data to the address of the entry table for indicating,
Including a method. The method of claim 1 , wherein the size of each entry in the entry table is 3 bits. The method according to claim 1 , wherein the pointer is composed of the N bits, and the N bits are lower-order N bits constituting the MAC address . The method of claim 1 , wherein the N bits are obtained by subtracting the entry table size from all the MAC address numbers . The method according to claim 1 , wherein the MAC pointer table includes a first pointer mask register for designating the N bits among bits constituting the MAC address . The MAC pointer table has a second pointer mask register for designating M bits excluding the N bits from among the bits constituting the MAC address, and the pointer is derived from (N + M) bits of the MAC address. It becomes the method of claim 5. A recording medium on which a computer-readable program for rearranging MAC addresses is recorded, the program being stored in the computer,
(A) generating an entry table of the MAC address having a predetermined number of entries determined from a table size derived from a MAC address search allowable time;
(B) generating a MAC pointer table having the number of N-bit entries in the bits constituting the MAC address;
(C) obtaining MAC address entry data;
(D) determining whether the pointer of each entry is valid or invalid from the bit information of each entry in the MAC pointer table;
(E) storing data at the address of the entry table indicated by the pointer determined to be valid in the MAC pointer table;
A recording medium that causes The recording medium according to claim 7, wherein the size of each entry of the MAC pointer table is 3 bits. The recording medium according to claim 7, wherein the pointer includes the N bits, and the N bits are lower-order N bits constituting the MAC address. The recording medium according to claim 7, wherein the N bits are obtained by subtracting the entry table size from all the MAC address numbers. The MAC pointer table specifies a first pointer mask register for designating the N bits from among the bits constituting the MAC address, and M bits excluding the N bits from the bits constituting the MAC address. The recording medium according to claim 7, further comprising: a second pointer mask register, wherein the pointer includes (N + M) bits of the MAC address.

Images