KR101787899B1 - A bitmap generating method based on priority-name prefix trie and name prefix searching method using the bitmap - Google Patents

Abstract

A method of generating a bitmap based on a priority name prefix trie and a method of searching for a name prefix using the bitmap are disclosed. Priority Name Prefix Trie-based bitmap generation methods include a Priority-Name Prefix Trie (P-TEF) including priority nodes and non-priority nodes derived from a Name Prefix Trie (NPT) NPT); Generating a bitmap including a bitmap table, an encoding table and a node table from the priority-name prefix trie; And storing the generated bitmap.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for generating a bitmap based on a priority name prefix trie and a method for searching for a name prefix using the bitmap,

The present invention relates to a method for generating a bitmap based on a priority name prefix trie and a method for searching for a name prefix using the bitmap. More specifically, the present invention relates to a priority node derived from a name prefix trie (NPT) (Bitmap table, encoding table and node table) generated from the determined priority-name prefix trie are determined based on the priority-name prefix trie (P-NPT) To retrieve a name prefix for an input packet.

The current Internet has been developed for the purpose of sharing resources based on TCP / IP. Therefore, such TCP / IP based Internet is suitable for data transmission between two end systems. Today, however, Internet traffic is not simply data transmission, but distribution and application of large-capacity contents such as web streaming, video transmission, peer-to-peer (P2P) Due to such changes in the Internet traffic pattern, the existing Internet shows limitations such as lack of bandwidth and deterioration of service quality. Named Data Network (NDN) is proposed as a form of future Internet to solve the problems of the Internet.

In existing Internet routers and NDN routers, packet transmission is performed based on the longest prefix. However, in the process of finding the longest matching prefix, the existing Internet router uses the IP address, but the NDN router uses the name of the content. Since the name of the content is not limited by the length and the format, unlike the IP address having the fixed length, a large amount of memory is required to implement the forwarding information base (FIB) table of the NDN router. For this reason, the FIB table of the NDN router can only be implemented using chip-off-chip memory.

However, the processing time using a chip-external memory such as DRAM is about 10 to 20 times the processing time using chip-internal memory such as SRAM. The process of finding the longest matching prefix using the FIB table stored in the chip-external memory is a major factor in lowering the overall packet transmission rate. In the router, packet transmission must be performed in accordance with the packet wire-speed, so that the reduction in the search speed in the FIB table is directly related to the performance degradation of the router.

The present invention determines a priority-name prefix trie (P-NPT) including a priority node and a non-priority node derived from a name prefix trie (NPT) The present invention provides an apparatus and method for improving a name prefix search capability by retrieving a name prefix for an input packet using a bit map (bit map table, encoding table and node table) generated from a name prefix trie.

A method for generating a bitmap based on a priority name prefix trie according to an embodiment of the present invention includes a priority-name prefix trie including a priority node and a non-priority node derived from a name prefix trie (NPT) Priority-Name Prefix Trie, P-NPT); Generating a bitmap including a bitmap table, an encoding table and a node table from the priority-name prefix trie; And storing the generated bitmap.

Wherein the determining step comprises: determining a subtree in which an empty node having no name prefix in the name prefix tri is set as a root node; Determining an insertion node storing a name prefix having the largest number of components among at least one insertion node having a name prefix in the subtree as a priority node and replacing the insertion node with a root node of the subtree.

Wherein the step of generating the node number of the priority node included in the priority-name prefix triangle and the node number of the non-priority node indicating the inserted node excluding the priority node is performed by using the index number of the bitmap table and the index number of the node table . ≪ / RTI >

The bitmap table includes (i) a node type indicating whether the nodes included in the priority-name prefix trie are priority nodes or non-priority nodes, (ii) a node type indicating whether nodes included in the priority-name prefix tri (Iii) a hash brief that indicates a fingerprint derived by applying a hash function to a name prefix stored in each of the nodes included in the priority-name prefix triad, (iii) a component representing the number of components constituting the stored name prefix, (iv) Child number information indicating the number of child nodes for the nodes included in the priority-name prefix trie, and (iv) Encoding corresponding to the path to the first child node of the nodes included in the priority- And at least one of the child pointer information indicating the address value of the table have.

Wherein the encoding table includes edge information for child nodes of the priority node and the non-priority node included in the priority-name prefix triangle, and edge information for the child nodes of the priority node and the non-priority node, Quot; < / RTI >

The node table may include a name prefix included in each of the priority node and the non-priority node of the priority-name prefix trie, and an output face corresponding to the name prefix.

The storing step may store the bitmap table and the encoding table in an internal memory of a chip that retrieves a name prefix, and the node table may store the name prefix in an external memory of a chip that retrieves a name prefix.

A method for searching for a name prefix according to an embodiment of the present invention includes: identifying a bitmap generated based on a priority-name prefix triangle including a priority node and a non-priority node derived from a name prefix triage; And searching for a name prefix in a priority-name prefix triangle in which the name component of the input packet matches the most components using the identified bitmap, wherein the bitmap includes a bitmap table, an encoding table, And a node table.

The searching step may include determining whether a node included in the priority-name prefix trie is a priority-node or a non-priority node based on a node type of the bitmap table accessed to search for a name prefix of the input packet ; And extracting an output face by accessing an index number of a node table identical to the index number of the verified non-priority node when the node included in the priority-name prefix tri is identified as a non-priority node .

Confirming whether the non-priority node of the bitmap table accessed to retrieve the name prefix of the input packet after the extracting has a child node; Accessing the encoding table using the child pointer information of the bitmap table when the non-priority node has determined that the child node has the child node, Checking edge information; An index number for a non-priority node of the bitmap table accessed to search for a name prefix of the input packet, and an encoding number corresponding to the identified edge information if the edge information for the matching child node is identified And determining an index of the bitmap table to be searched next.

Wherein the searching step comprises: checking whether a node included in the priority-name prefix trie is a priority node or a non-priority node based on a node type of the bitmap table accessed to search for a name prefix of the input packet ; Comparing the hash value of the name prefix of the input packet with the hash brief included in the bitmap table if the node included in the priority-name prefix triangle is identified as a priority node; If a hash value of a name prefix of the input packet matches a hash brief included in the bitmap table, accessing an index number of a node table that is the same as an index number of the bitmap table, Comparing a name prefix stored in the table with a name prefix; And extracting an output face if the name prefix of the input packet and the name prefix stored in the node table coincide with each other.

Wherein the priority-name prefix trie determines a subtree setting an empty node having no name prefix to the root node in the name prefix triad, and wherein in the subtree the largest number of at least one insertion node having a name prefix , And replacing the insertion node storing the name prefix having the name prefix with the root node of the subtree as a priority node.

The bitmap may include a priority node included in the priority-name prefix trie and a node number of a non-priority node indicating an insertion node excluding the priority node to an index number of the bitmap table and an index number of the node table As shown in FIG.

The bitmap table includes (i) a node type indicating whether the nodes included in the priority-name prefix trie are priority nodes or non-priority nodes, (ii) a node type stored in each of the nodes included in the priority- (Iii) a hash brief that indicates a fingerprint derived by applying a hash function to a name prefix stored in each of the nodes included in the priority-name prefix trie; (iii) a component indicating the number of components constituting the name prefix; iv) a child number information indicating the number of child nodes for the nodes included in the priority-name prefix trie, and (iv) an encoding table corresponding to the path to the first child node of the nodes included in the priority- And child pointer information indicating an address value of the child pointer information .

Wherein the encoding table includes edge information for child nodes of the priority node and the non-priority node included in the priority-name prefix triangle, and edge information for the child nodes of the priority node and the non-priority node, Quot; < / RTI >

The node table may include a name prefix included in each of the priority node and the non-priority node of the priority-name prefix trie, and an output face corresponding to the name prefix.

The bitmap table and the encoding table are stored in an internal memory of a chip retrieving a name prefix, and the node table can be stored in an external memory of a chip retrieving a name prefix.

According to an embodiment of the present invention, a priority-name prefix trie (P-NPT) including a priority node and a non-priority node derived from a name prefix trie (NPT) And search for the name prefix for the input packet using the bitmap (bitmap table, encoding table, and node table) generated from the determined priority-name prefix trie to improve the name prefix search capability.

According to an embodiment of the present invention, a bitmap (bitmap table, encoding table, and node table) generated from a priority-name prefix trie is divided and stored in an internal memory and an external memory of a chip that searches for a name prefix, The search ability can be improved.

1 is a diagram illustrating a name prefix search apparatus according to an embodiment of the present invention.
2 is an exemplary set of name prefixes and an example of a name prefix trie constructed using the example set in accordance with an embodiment of the present invention.
3 is a diagram illustrating an example of a priority name prefix triangle according to an embodiment of the present invention.
FIG. 4 is a diagram illustrating an example of a bitmap based on a priority name prefix triangle according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating a method of searching for a name prefix for a node of a priority name prefix triad determined to be nonpriority according to an exemplary embodiment of the present invention.
FIG. 6 is a diagram illustrating a method of searching for a name prefix for a node of a priority name prefix triage determined to have a priority according to an embodiment of the present invention.

It is to be understood that the specific structural or functional descriptions of embodiments of the present invention disclosed herein are presented for the purpose of describing embodiments only in accordance with the concepts of the present invention, May be embodied in various forms and are not limited to the embodiments described herein.

Embodiments in accordance with the concepts of the present invention are capable of various modifications and may take various forms, so that the embodiments are illustrated in the drawings and described in detail herein. However, it is not intended to limit the embodiments according to the concepts of the present invention to the specific disclosure forms, but includes changes, equivalents, or alternatives falling within the spirit and scope of the present invention.

The terms first, second, or the like may be used to describe various elements, but the elements should not be limited by the terms. The terms may be named for the purpose of distinguishing one element from another, for example without departing from the scope of the right according to the concept of the present invention, the first element being referred to as the second element, Similarly, the second component may also be referred to as the first component.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between. Expressions that describe the relationship between components, for example, "between" and "immediately" or "directly adjacent to" should be interpreted as well.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms "comprises ", or" having ", and the like, are used to specify one or more of the features, numbers, steps, operations, elements, But do not preclude the presence or addition of steps, operations, elements, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the meaning of the context in the relevant art and, unless explicitly defined herein, are to be interpreted as ideal or overly formal Do not.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. Like reference symbols in the drawings denote like elements.

1 is a diagram illustrating a name prefix search apparatus according to an embodiment of the present invention.

The name prefix search apparatus 100 may include a processor 110, an internal memory 120 disposed inside the processor, and an external memory 130 disposed outside the processor. At this time, the processor 110 may be implemented as one chip. Thus, if the processor 110 utilizes the internal memory 120 disposed within the processor 110 to retrieve the name prefix for the input packet, the name prefix search may proceed on-chip have. However, if the processor 110 uses the external memory 130 located outside the processor 110 to retrieve the name prefix, the name prefix search may be performed off-chip.

At this time, the search for the name prefix in the off-chip is slower than the searching for the name prefix in the on-chip due to the access time to the external memory 130. That is, searching for a name prefix on the off-chip has the greatest effect on the speed of searching for a name prefix because it takes a long time to use it, so the number of off-chip accesses is used as an index of performance evaluation for a name prefix search .

Accordingly, the name prefix search apparatus 100 may divide and store the FIB table in the internal memory 120 and the external memory 130 of the processor 110 to reduce the time taken to search for a name prefix for an input packet.

Specifically, the name prefix search apparatus 100 identifies a bitmap generated based on a priority-name prefix triangle including a priority node and a non-priority node derived from a name prefix triage, and based on the identified bitmap The name prefix for the input packet can be retrieved. That is, the name prefix search apparatus 100 stores the bitmap table and the encoding table including the outline information of the priority-name prefix trie among the bitmaps in the internal memory 120, The processing in the external memory 130 can be minimized by storing the name prefix and the output face, which are substantial information on each node, in the external memory 130. [ At this time, a method of determining the priority-name prefix trie and a method of generating the bitmap will be described later in detail.

2 is an exemplary set of name prefixes and an example of a name prefix trie constructed using the example set in accordance with an embodiment of the present invention.

The Name Prefix Trie (NPT) is an extension of the binary trie used in IP address lookup. Unlike an IP address that uses only 0s and 1s as path values, it is used as a path value in an NDN (Named Data Network), since the path to the node where the prefix is stored means the prefix. Since the name of the content is not limited in length and form, the number of child nodes that a node can have is infinitely large. In this case, since the name prefix trie is an extension of the binary trie used in the IP address search, like the binary trie, there is a lot of memory waste due to an unnecessary internal node. In addition, there is a problem in that the search performance is poor because the search is to be continued until there is no more path to search for the name prefix of the input packet and the name prefix having the largest number of components.

As a solution to this problem, the present invention provides a method for determining a priority-name prefix trie (P-NPT) including a priority node and a non-priority node derived from a name prefix trie.

2 (a) is an example set of a name prefix and an output face, and FIG. 2 (b) shows an example of building a name prefix triangle for FIG. 2 (a). Hereinafter, a node in which the name prefix is stored in the name prefix trie is referred to as an insertion node, and an empty node inevitably exists in the path to the node where the name prefix is stored. At this time, each name prefix has output face information, and the insertion node of the name prefix triad can also be set to a different node number.

For example, if you build a name prefix kr / or / visitkorea with a name prefix trie: First, the empty node 210 is located at the highest level of the name prefix triad. Thereafter, when the user moves along the path corresponding to kr from the empty node 210, the empty node 220 is located. If the user moves along the path corresponding to or from the empty node 220, the empty node 230 is located. If the empty node 230 moves along the path corresponding to the visit korea from the empty node 230, the insertion of the name prefix kr / or / Node 240 may be located.

3 is a diagram illustrating an example of a priority name prefix triangle according to an embodiment of the present invention.

Unlike a name prefix trie that only had an insert node and an empty node, the priority-name prefix trie determined from the name prefix trie is a priority node and a non-priority node representing the inserted node excluding the priority node node.

The process of determining the priority-name prefix trie is as follows. First, in the name prefix trie, an empty node that does not have a name prefix can be determined as a subtry that is set as the root node. The empty node selected at this time may be a node existing at the highest level of the name prefix trie. Thereafter, the insertion node storing the name prefix having the largest number of components among at least one insertion node in which the name prefix is stored in the determined subtree may be determined as the priority node and replaced with the root node of the subtree. The process of determining such a priority-name prefix triage may be performed by the processor 110 included in the name prefix search apparatus 100. [

For example, a process of determining a name-prefix trie as shown in FIG. 2 (b) as a priority-name prefix trie is as follows. First, the processor 110 may determine a sub-trie 311 that sets an empty node 310 located at the highest level of the name prefix triangle as shown in FIG. 3A as a root node. After that, the processor 110 may determine the insertion node 312 storing the name prefix having the largest number of the insertion nodes existing in the subtree 311 as the priority node. At this time, the name prefix stored in the insertion node 312 determined as the priority node is 'com / youtube / user / skyDoesMinecraft' and the output face is 7. The processor 110 may substitute such a name prefix and output face in the root node 310 of the subtree 310 for storage. At this time, when there are two or more insertion nodes 312 and 313 in which a name prefix having the largest number of components is stored in the same subtree 311, the processor 110 may select either one of them and replace it with a root node. In the present invention, when there are more than two insertion nodes 312 and 313 storing a name prefix having the largest number of components, the insertion node 312 located on the left side is determined as a priority node and replaced with a root node.

You can now jump to the next lower level because there are no more empty nodes at the top level of the name prefix trie. Thus, the processor 110 may determine the subtree 321 where the empty node 320, which is in the second level of the name prefix triad, is set as the root node. Then, the processor 110 may determine the insertion node 322 storing the name prefix having the largest number of the insertion nodes existing in the subtree 321 as a priority node. At this time, the name prefix stored in the insertion node 322 determined as the priority node is 'kr / or / visitkorea', and the output face is 1. Processor 110 may replace and store such name prefix and output face in root node 320 of subtree 321. Similarly, if there are more than two insertion nodes 322 and 323 storing the name prefix having the largest number of components in the same subtree 321, the processor 110 may select one of them and replace it with the root node.

The processor 110 may build the priority-name prefix trie as shown in FIG. 3 (b) by repeating this process until there are no more empty nodes in the name prefix trie. In this way, the priority-name prefix trie stores the information of the inserted node which is likely to match most components when the priority node is accessed to retrieve the name prefix. Accordingly, if the name prefix of the input packet matches the name prefix information stored in the priority node, the processor 110 can terminate the search even if there is a path to continue the search. That is, by defining the priority node through such priority-name prefix trie, it is possible to reduce unnecessary empty nodes and improve search performance.

FIG. 4 is a diagram illustrating an example of a bitmap based on a priority name prefix triangle according to an embodiment of the present invention.

Each node of the name prefix trie has a flag (1 bit), a pointer to a child node, edge information (32 bits per edge) for a child node, a name prefix (128 bits) It requires a lot of memory. In the present invention, the outline information of the priority-name prefix tri is stored in the internal memory 120 of the processor 110 by applying a bitmap, and the outline information of the priority-name prefix tri is stored in the external memory 130 disposed outside the processor 110 Can be minimized. Such a bitmap may be generated by the processor 110 of the name prefix searching apparatus 100 based on the priority-name prefix trie. The bitmap example of FIG. 4 was generated based on the priority-name prefix trie of FIG.

 The bitmap may consist of a bitmap table, an encoding table, and a node table. The bitmap table and encoding table are then stored in the internal memory 120 of the processor 110 that retrieves the name prefix and the node table may be stored in the external memory 130 of the processor 110 retrieving the name prefix have.

The node number of the priority-name prefix trie starts from 0 and can be assigned in order from the highest level to the lowest level. For example, the node number of the root node in the priority-name prefix triangle shown in FIG. 3 (b) is 0, and the node number of the priority node storing 'kr / or / warning' is 5.

In generating the bitmap using the priority-name prefix trie, the index number of the bitmap table and the index number of the node table, which are the same as the respective node numbers included in the priority-name prefix trie, are matched, Can be stored. Therefore, there is no need for a separate pointer to connect the bitmap table and the node table.

A. Specifically, the bitmap table constituting the bitmap is composed of five pieces of field information for each node included in the priority-name prefix trie. First, as shown by column a in the bitmap table of FIG. 4, information on the node type indicating whether each of the nodes included in the priority-name prefix trie is a priority node or a non-priority node may be included. That is, if the node of the priority-name prefix trie that is accessed to retrieve the name prefix of the input packet is an insertion node, 0 is stored, and when the node is a priority node, 1 is stored.

Next, the bitmap table may include information on the number of components of the name prefix stored in each of the nodes included in the priority-name prefix trie, such as column b. For example, index 0 of the bitmap table represents a priority node at the highest level of the priority-name prefix trie in FIG. Node 0 of the priority-name prefix trie stores the name prefix of 'com / youtube / user / skyDoesMinecraft', and each com, youtube, user, and skyDoesMinecraft correspond to the components that make up the name prefix. Thus, the number of components corresponding to index 0 in the bitmap table is four.

The bitmap table may then include a hash brief that is derived by applying a hash function to the name prefix stored in each of the nodes included in the priority-name prefix triad, such as column c. At this time, the derived hash brief may indicate a fingerprint for the name prefix.

Such a hash brief can be used when the name prefix search device 100 searches for a name prefix. In the name prefix trie, the name prefix for the input packet matches all the name prefixes present in the path, while the priority prefix of the priority trie stores the name prefix with the largest number of components. It is less likely to match the name prefix stored in the priority node. Due to the characteristics of the priority-name prefix triage, unnecessary access to the external memory 130 may occur frequently. In the present invention, the concept of a hash brief is proposed and applied to determine a hash value for a component of a name prefix for an input packet corresponding to the number of components stored in a priority node, and access to the external memory 130 You can decide. Namely, since the determined hash value is compared with the hash brief stored in the bitmap table and only the matching is performed, the access to the external memory 130 is performed. Therefore, the access to the external memory 130, which is unnecessarily generated due to the characteristics of the priority- Access can be minimized.

Next, the bitmap table may include child number information indicating the number of child nodes for each of the nodes included in the priority-name prefix trie as column d. For example, node 0 of the priority-name prefix trie has four child nodes 1, 2, 3, and 4 at the bottom. Therefore, the child number information corresponding to the index 0 of the bitmap table is 4.

The bitmap table may then include child pointer information for each of the nodes included in the priority-name prefix trie, such as column e. The child pointer information may indicate the address value of the encoding table corresponding to the path to the first child node of each of the nodes included in the priority-name prefix trie.

B. The encoding table constituting the bitmap includes edge information for the child nodes of each of the nodes (priority node and non-priority node) included in the priority-name prefix trie, and edge information for the priority-name prefix tri And an encoding number indicating an index number difference between the included nodes and the child node.

The number of child nodes in the prefix search using the IP address is limited to a maximum of 2, but the number of child nodes in the NDN is not limited. Therefore, the present invention introduces an encoding table to solve the difficulty in applying a bitmap considering the number of unspecified child nodes for each of the nodes included in the priority-name prefix trie. If the bitmap table stores the outline information of the nodes included in the priority-name prefix trie, the encoding table may contain edge information to the child nodes of the nodes included in the priority-name prefix trie . In addition, the encoding table can link the node of the priority-name prefix trie with the child node using the encoding number indicating the node number difference between the node and the child node included in the priority-name prefix trie.

For example, node 0 of the priority-name prefix trie has four child nodes 1, 2, 3, and 4 in the lower level. At this time, the edge information corresponding to the path from the node 0 to each child node is kr, de, hk, and com, and the edge information is sequentially stored in the encoding table. As can be seen from FIG. 3 (b), when the node 0 moves along the edge information kr, the node # 1 of the priority-name prefix trie is searched as a child node. Therefore, the encoding table can store 1, which is the node number difference between node 0 and node 1, as an encoding number. Similarly, if node 0 moves along the edge information de, node 2 of the priority-name prefix trie is searched as a child node. At this time, the encoding table may be stored with a binary encoding number, which is a number difference between nodes 0 and 2. [

C. The node table constituting the bitmap may store a name prefix included in each of the nodes (priority node and non-priority node) included in the priority-name prefix triangle and an output face corresponding to the name prefix .

The present invention can generate the bitmaps disclosed in A, B, and C using the priority-name prefix trie. Specifically, an example of generating the node # 4 included in the priority-name prefix trie of FIG. 3 as a bitmap is as follows. First, the node information of node 4 is stored in index 4 of the bitmap table and the node table. Since the node 4 is a priority node, the information about the node type is 1, and the stored name prefix is 'com / youtube / user / PewDiePie', so the component information is 4 (com, youtube, user, PewDiePie). Since the node 4 has four child nodes at the bottom, the child number information is 4. The edge information corresponding to the route from the node # 4 to the child node is empowernetwork, youtube, goole, and facebook in order, and these edge information is sequentially stored in the encoding table. And the child pointer of the bitmap table stores the address value of the encoding table corresponding to the path to the first child node (empowernetwork) of node 4. The node number difference between node 4 and node 6, which is the first child node, is stored as a 2-encoded number.

FIG. 5 is a diagram illustrating a method of searching for a name prefix for a node of a priority name prefix triad determined to be nonpriority according to an exemplary embodiment of the present invention.

The name prefix search apparatus 100 can identify the bit map generated based on the priority-name prefix triad including the priority node and the non-priority node derived from the name prefix triage. At this time, the identified bitmap may be composed of a bitmap table, an encoding table, and a node table.

First, in step 510, the name prefix search device 100 can access the nodes in the priority-name prefix tri based on the index number of the bitmap table to retrieve the name prefix of the input packet. At this time, the index number of the bitmap table may correspond to the node number in the priority-name prefix triangle. That is, the index 0 of the bitmap table can correspond to the node 0 in the priority-name prefix trie. Then, the name prefix search apparatus 100 can retrieve the name prefix of the input packet from the node at the highest level in the priority-name prefix triad.

Therefore, the name prefix search device 100 can confirm whether the node in the priority-name prefix trie is a priority node or a non-priority node based on the node type information included in index 0 of the bitmap table.

If it is determined in step 520 that the node included in the priority-name prefix trie is identified as a non-priority node, the name prefix search apparatus 100 determines that the index number of the node table is the same as the index number of the verified non- It is accessible. Then, the name prefix search device 100 can extract and store the output face stored in the index number of the accessed node table.

Thereafter, in step 530, the name prefix search apparatus 100 can confirm whether the verified non-priority node has a child node by using the child number information of the bitmap table.

If the identified non-priority node does not have any child nodes, the name prefix search device 100 may return the output face stored in step 520 as a search result as in step 560 and terminate the search .

In contrast, if the identified non-priority node has child nodes, the name prefix search apparatus 100 uses the child pointer information of the bitmap table corresponding to the identified non-priority node in step 540, It is accessible. The name prefix search apparatus 100 can then use the encoding table to identify edge information for child nodes that match the path to be searched next.

At this time, if there is no edge information for the matching child node, the name prefix search device 100 returns the search result as the output face stored in the step 520 as in the step 560, and terminates the search .

However, if there is edge information for the matching child node, the name prefix search device 100 uses the index number of the verified non-priority node and the encoding number corresponding to the identified edge information at step 550, The index of the bitmap table to be searched can be determined.

The name prefix search device 100 may then retrieve the name prefix for the input packet by repeating steps 510 through 560 based on the index of the determined bitmap table.

FIG. 6 is a diagram illustrating a method of searching for a name prefix for a node of a priority name prefix triage determined to have a priority according to an embodiment of the present invention.

First, in step 610, the name prefix search device 100 can access the node in the priority-name prefix tri based on the index number of the bitmap table to retrieve the name prefix of the input packet. At this time, the index number of the bitmap table may correspond to the node number in the priority-name prefix triangle. That is, the index 0 of the bitmap table can correspond to the node 0 in the priority-name prefix trie. Then, the name prefix search apparatus 100 can retrieve the name prefix of the input packet from the node at the highest level in the priority-name prefix triad.

Therefore, the name prefix search device 100 can confirm whether the node in the priority-name prefix trie is a priority node or a non-priority node based on the node type information included in index 0 of the bitmap table.

In step 620, the name prefix search device 100 may compare the hash value of the hash brief for the priority node identified in step 610 and the hash value for the name prefix of the input packet. At this time, the hash value for the name prefix of the input packet can be extracted using the hash function that is the same as the hash function used for generating the hash brief of the bitmap table. The name prefix search apparatus 100 may determine a hash value based on a component constituting a name prefix for an input packet corresponding to the number of components of the name prefix stored in the identified priority node.

At this time, if the hash brief for the compared priority node and the hash value for the name prefix of the input packet coincide with each other, the name prefix search apparatus 100 searches for the name prefix stored in the identified priority node in step 630 It is possible to compare whether the index number of the identified priority node matches the name prefix stored in the index number of the same node table.

If the comparison result shows coincidence, the name prefix search apparatus 100 may extract and store the output face stored in the index number of the node table identical to the index number of the identified priority node, as in step 640.

Thereafter, in step 680, the name prefix search device 100 may return the output face stored in step 640 to the search result and terminate the search.

If the hash value for the compared priority node and the hash value for the name prefix of the input packet do not match, the name prefix search device 100 uses the child number information of the bitmap table in step 650 It is possible to confirm whether or not the identified priority node has child nodes.

If the identified priority node does not have a child node, the name prefix search device 100 may return the search result of the output face stored in step 640 as in step 680 and terminate the search .

However, if the identified priority node has a child node, the name prefix search apparatus 100 accesses the encoding table using the child pointer information of the bitmap table corresponding to the identified priority node in step 660 can do. The name prefix search apparatus 100 can then use the encoding table to identify edge information for child nodes that match the path to be searched next.

At this time, if there is no edge information for the matching child node, the name prefix search device 100 returns the search result as the output face stored in the step 640 as in the step 680, and terminates the search .

However, if there is edge information for the matching child node, the name prefix search device 100 uses the index number of the identified priority node and the encoding number corresponding to the identified edge information at step 670 The index of the bitmap table to be searched can be determined.

The name prefix search apparatus 100 may then retrieve the name prefix for the input packet by repeating steps 610 through 680 based on the index of the determined bitmap table.

For example, according to the name prefix search method, searching for the input packet's name prefix 'com / goole / mail / mailbox / user1' First, the node of the priority-name prefix triangle corresponding to the index 0 of the bitmap table corresponds to the priority node because the node type is 1. The name prefix search apparatus 100 can determine a hash value using only 'comgoolemailmailbox' among 'com / goole / mail / mailbox / user1' because the number of components corresponding to index 0 is 4. At this time, suppose that the determined hash value is 010.

Since the hash value determined at this time does not match the hash brief corresponding to the index 0, the name prefix search apparatus 100 can access the encoding table using the child pointer information. The name prefix search apparatus 100 can then use the encoding table to identify edge information for child nodes that match the path to be searched next. As a result, it is confirmed that 'com' in the name prefix of the input packet matches.

The name prefix search apparatus 100 adds the encoding number 4 corresponding to 'com' in the encoding table and the index number 0 to the node of the priority-name prefix trie in the current search, Number 4 can be determined.

Thereafter, the name prefix search device 100 can re-search the name prefix based on the index number 4 of the newly determined bitmap table. The node of the priority-name prefix triangle corresponding to the index number 4 of the bitmap table is the priority node, so the hash value for the name prefix of the input packet can be compared with the hash brief corresponding to the index No. 4.

If the comparison result does not match, the name prefix search device 100 can access the encoding table using the child pointer information. The name prefix search apparatus 100 can then use the encoding table to identify edge information for child nodes that match the path to be searched next. As a result, we can confirm that 'google' matches the name prefix of the input packet.

Then, the name prefix search apparatus 100 adds the encoding number 4 corresponding to 'google' in the encoding table and the index number 4 to the node of the priority-name prefix trie in the current search, The index number 8 can be determined.

Then, the name prefix search device 100 can re-search the name prefix based on the index number 8 of the determined bitmap table. The node of the priority-name prefix trie corresponding to index number 8 of the bitmap table is a non-priority node. Therefore, the name prefix search apparatus 100 can access the index number of the node table that is the same as the index number 8 of the bitmap table. Then, the name prefix search apparatus 100 can extract and store the output face 9 stored in the index number of the accessed node table.

At this time, since the priority-name prefix trie corresponding to the index number 8 of the bitmap table has no child node, the name prefix search device 100 can return the stored output face 9 and terminate the search.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA) , A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

100: Name prefix search device
110: Processor
120: Internal memory
130: External memory

Claims (17)

A priority name prefix triage based bit map generation method performed by a name prefix search apparatus,
The processor of the name prefix searching apparatus determines a priority-name prefix trie (P-NPT) including priority nodes and non-priority nodes derived from the Name Prefix Trie (NPT) step;
Wherein the processor of the name prefix searching device generates a bitmap including the bitmap table, the encoding table and the node table from the priority-name prefix trie; And
The processor of the name prefix retrieval apparatus stores the generated bitmap
A method for generating a bitmap based on a priority name prefix trie. The method according to claim 1,
Wherein the determining comprises:
The processor of the name prefix search apparatus determining a subtree in which an empty node having no name prefix in the name prefix tri is set as a root node; And
The processor of the name prefix searching apparatus determines an insertion node storing a name prefix having the largest number of components among at least one insertion node having a name prefix in the subtree as a priority node and replacing the insertion node with a root node of the subtree
A method for generating a bitmap based on a priority name prefix trie. The method according to claim 1,
Wherein the generating comprises:
The processor of the name prefix search apparatus searches for the node number of the priority node included in the priority-name prefix trie and the non-priority node indicating the insertion node excluding the priority node as the index number of the bitmap table and the index A priority name that matches each of the numbers. A prefix trie based bitmap generation method. The method according to claim 1,
Wherein the bitmap table comprises:
(i) a node type indicating whether the nodes included in the priority-name prefix trie are priority nodes or non-priority nodes,
(ii) a component representing the number of components constituting the name prefix stored in each of the nodes included in the priority-name prefix trie,
(iii) a hash brief that indicates a fingerprint derived by applying a hash function to a name prefix stored in each of the nodes included in the priority-name prefix triad,
(iv) Child number information indicating the number of child nodes for the nodes included in the priority-name prefix trie and
(iv) Priority name - A priority name that includes at least one of the child pointer information indicating the address value of the encoding table corresponding to the path to the first child node of the nodes included in the prefix trie. Generation method. The method according to claim 1,
Wherein the encoding table comprises:
Edge information for the child node of each of the priority node and the non-priority node included in the priority-name prefix triangle and an encoding number indicating the difference between the index numbers of the priority node, the non-priority node, and the child node Included priority names Prefix trie based bitmap generation methods. The method according to claim 1,
The node table comprising:
A priority name prefix tri based bitmap generation step including a name prefix included in each of a priority node and a non-priority node of the priority-name prefix trie, and an output face corresponding to the name prefix. The method according to claim 1,
Wherein the storing step comprises:
The processor of the name prefix retrieval apparatus stores the bitmap table and the encoding table in the internal memory of the chip retrieving the name prefix,
Wherein the node table stores a name prefix in an external memory of a chip that retrieves a priority name prefix tri based bitmap. The name prefix search method performed by the name prefix search device includes:
Identifying a bitmap generated based on a priority-name prefix triangle including a priority node and a non-priority node derived from a name prefix trie of a name prefix search apparatus;
The processor of the name prefix searching apparatus searches for a name prefix in a priority-name prefix triangle in which the component name matches the name prefix of the input packet using the identified bitmap
Lt; / RTI >
The bit map may include:
A name prefix search method comprising a bitmap table, an encoding table, and a node table. 9. The method of claim 8,
Wherein the searching comprises:
The node included in the priority-name prefix trie based on the node type of the bitmap table accessed by the processor of the name prefix searching apparatus to search for the name prefix of the input packet is a priority- ; And
If the processor of the name-prefix search apparatus determines that the node included in the priority-name prefix trie is a non-priority node, the processor retrieves the output face by accessing the index number of the node table that is the same as the index number of the verified non- step
≪ / RTI > 10. The method of claim 9,
After the extracting step,
Confirming whether a non-priority node of the bitmap table that the processor of the name prefix search apparatus has accessed to search for a name prefix of the input packet has a child node;
The processor of the name prefix searching apparatus accesses the encoding table using the child pointer information of the bitmap table when the non-priority node has a child node, Identifying edge information for a child node matching the child node; And
Wherein when the processor of the name prefix searching apparatus determines that the edge information for the matching child node has been checked, the index number of the non-priority node of the bitmap table accessed to search for the name prefix of the input packet, Determining an index of a bitmap table to be retrieved next using an encoding number corresponding to the information
And a prefix search method. 9. The method of claim 8,
Wherein the searching comprises:
The processor of the name prefix searching apparatus determines whether the node included in the priority-name prefix trie is a priority node or a non-priority node based on the node type of the bitmap table accessed by the processor to search for a name prefix of the input packet ;
If the processor of the name prefix search apparatus determines that the node included in the priority-name prefix trie is a priority node, the hash value of the name prefix of the input packet is compared with the hash brief included in the bitmap table ;
The processor of the name prefix searching apparatus compares the hash value of the input packet with the hash value of the name prefix of the input packet and the hash brief included in the bit map table to see if the name prefix of the input packet matches the name prefix stored in the node table ; And
If the processor of the name prefix search apparatus matches the name prefix of the input packet and the name prefix stored in the node table, the index number of the node table that is the same as the index number of the node included in the identified priority- Step of accessing and extracting the output face
≪ / RTI > 9. The method of claim 8,
Wherein the priority-name prefix trie comprises:
Determining a subtree for setting an empty node having no name prefix in the name prefix triangle as a root node, and determining a subtree in which a name prefix having a name component Is determined as a priority node and replaced with the root node of the subtree. 9. The method of claim 8,
The bit map may include:
A priority node included in the priority-name prefix trie and a node number of a non-priority node indicating an insertion node excluding the priority node are matched with index numbers of the bitmap table and index numbers of the encoding table, respectively How to search for name prefixes. 9. The method of claim 8,
The bitmap table,
(i) a node type indicating whether the nodes included in the priority-name prefix trie are priority nodes or non-priority nodes,
(ii) a component representing the number of components constituting the name prefix stored in each of the nodes included in the priority-name prefix trie,
(iii) a hash brief that indicates a fingerprint derived by applying a hash function to a name prefix stored in each of the nodes included in the priority-name prefix triad,
(iv) Child number information indicating the number of child nodes for the nodes included in the priority-name prefix trie and
and (iv) child pointer information indicating an address value of an encoding table corresponding to a path to a first child node of nodes included in the priority-name prefix trie. 9. The method of claim 8,
Wherein the encoding table comprises:
Edge information for the child node of each of the priority node and the non-priority node included in the priority-name prefix trie and an encoding number indicating the difference between the index numbers of the priority node, the non-priority node and the child node Contains the name prefix search method. 9. The method of claim 8,
The node table comprising:
A name prefix included in each of a priority node and a non-priority node of the priority-name prefix triangle, and an output face corresponding to the name prefix. 9. The method of claim 8,
The bitmap table and the encoding table are stored in an internal memory of a chip which searches for a name prefix,
Wherein the node table is stored in an external memory of a chip retrieving a name prefix.

Images