CN104090942A - Trie search method and device applied to network processor - Google Patents
Trie search method and device applied to network processor Download PDFInfo
- Publication number
- CN104090942A CN104090942A CN201410307241.0A CN201410307241A CN104090942A CN 104090942 A CN104090942 A CN 104090942A CN 201410307241 A CN201410307241 A CN 201410307241A CN 104090942 A CN104090942 A CN 104090942A
- Authority
- CN
- China
- Prior art keywords
- tree
- trie
- node
- value
- indicate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000013500 data storage Methods 0.000 claims abstract description 24
- 230000008520 organization Effects 0.000 claims description 22
- 238000010276 construction Methods 0.000 description 7
- 238000013517 stratification Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 3
- 241001582888 Lobus Species 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000005055 memory storage Effects 0.000 description 2
- 206010006895 Cachexia Diseases 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 208000016318 wasting Diseases 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9027—Trees
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a Trie searching method applied to a network processor, which is characterized by comprising the following steps: step 1: establishing a multi-level storage structure, wherein the storage structure comprises a register file with gradually reduced priority, an internal RAM, an external SRAM and an external SDRAM; step 2: establishing a balanced tree, specifically, traversing each bit of all routing tables, finding out a bit with the number of values 0 and 1 being basically equal, setting the bit as a current root node, marking the bit as used, and circularly establishing a relative balanced tree; and step 3: and performing data storage, wherein for each data storage, adjacent bits of data are organized into a sub-tree structure for data storage. The Trie structure adopts a hierarchical storage structure, reduces the access time of a key to an external memory, and obtains a faster search speed with lower memory cost.
Description
Technical field
The present invention relates to network processing unit, particularly, relate to a kind of Trie search method being applied in network processing unit.
Background technology
Network processing unit is due to the good core devices that becomes next generation network product of fast, the able to programme performance of its speed, along with network demand improves constantly, the linear speed of network processing unit requires more and more higher, the package forward speed requiring is also more and more higher, and this makes speed of route lookup become the Main Bottleneck of restriction network processing unit performance.
The Routing Algorithm adopting in during network core is numerous, wherein, algorithm based on Trie tree not only has good seek rate, space complexity and time complexity, and can improve constantly router performance, therefore novel modern Fast IP-Routing Lookup Algorithm is all to realize by optimized algorithm on the basis of Trie tree, such as Trie tree path compression algorithm, Multi-branch Trie Tree algorithm etc.Path compression Trie tree algorithm is the same with binary tree algorithm, needs a large amount of memory access operation in its search procedure; And Multi-branch Trie Tree algorithm adopt be greater than 1 search step wide, improved memory access speed, but bandwidth of memory utilization factor is not high.
Summary of the invention
For defect of the prior art, a kind of method that the object of this invention is to provide fast finding, high memory bandwidth utilization factor realizes the searching structure that Trie searches.Technical matters to be solved by this invention:
1, according to the requirement of network processing unit linear speed, key word is searched for, Frame can find result in time, then correctly carries out processing data packets according to lookup result;
2, in search procedure, the memory device of stratification is provided, comprise register file, internal RAM, external SRAM and tetra-kinds of memory hierarchys of outside DDR SDRAM, make the storage resources of this Trie search structure low-cost realize search speed fast, the time cost of searching is each time about once the outside DDR SDRAM access time;
3, in contributing, make each branch's relative equilibrium can reduce memory access number of times, reduce the memory access time of worst case by the binary tree of balance;
4, the multiple node organizations in the balanced binary tree construction of the 3rd are become to a stalk tree construction, be stored in a storage block, make each time accessing operation can carry out multidigit and search, make full use of the feature of current storage high bandwidth, reduce memory access number of times, improve search efficiency.
A kind of Trie search method being applied in network processing unit providing according to an aspect of the present invention, comprises the steps:
Step 1: set up multi-level storage organization, described storage organization comprises register file, internal RAM, external SRAM and the outside SDRAM that priority diminishes gradually;
Step 2: equilibrium establishment tree, be specially, travel through each of all routing tables, find out the immediate position of number of its intermediate value 0 and value 1, this is set to current root node, and this position is labeled as and is used, circulate to realize the foundation of relative equilibrium tree with this;
Step 3: carry out data storage, wherein, to data storage each time, the sub-tree structure all contiguous several bit data being organized into is carried out data storage.
Preferably, the storage mode of described storage organization is divided into: the root node of Trie is placed in register file; The 2nd to 4 layers of Trie structure are put in internal RAM; The data that Trie structure are greater than to the 4th layer are placed in external SRAM or SDRAM according to the size of resource.
Preferably, described step 2 comprises the steps:
Step 2.1: according to all data strip destination data positions, if the number of the value of n bit data 0 and value 1 is equal, n position is arranged to the root node of tree;
Step 2.2: differ less according to the value 0 in m position and the number of value 1, so m position is arranged to left branch or the right branch of root node.
Preferably, described balanced tree comprises several sub-tree structure, and the organizational form of subtree in storer is: 0th～4 represent the topological structure of this subtree, and it can carry out structural adjustment according to memory bandwidth requirements; 5th～13 indicate the figure place that in subtree topological structure, first node is searched; 59th～67 in order to indicate the figure place of last node representative in subtree topological structure; Whether 68th～75 be leaf node in order to indicate searched next stage; 76th～91 in order to indicate the storer at place; 95th～255 in order to indicate next stage address of node, and successively, every 20 show a node link address.
A kind of Trie search device being applied in network processing unit providing according to another aspect of the present invention, comprises as lower device:
Storage organization apparatus for establishing, for setting up multi-level storage organization, described storage organization comprises register file, internal RAM, external SRAM and the outside SDRAM that priority diminishes gradually;
Balanced tree apparatus for establishing, for equilibrium establishment tree, is specially, travel through each of all routing tables, find out immediate of the number of its intermediate value 0 and value 1, this is set to current root node, and this position is labeled as and is used, circulate to realize the foundation of relative equilibrium tree with this;
Data storage device, for carrying out data storage, wherein, to data storage each time, the sub-tree structure all contiguous several bit data being organized into is carried out data storage.
Preferably, the storage mode of described storage organization is divided into: the root node of Trie is placed in register file; The 2nd to 4 layers of Trie structure are put in internal RAM; The data that Trie structure are greater than to the 4th layer are placed in external SRAM or SDRAM according to the size of resource.
Preferably, balanced tree apparatus for establishing, first according to all data strip destination data positions, is found out the value 0 of n bit data equal with the number of value 1, n position is arranged to the root node of tree; Secondly differ less according to the value 0 in m position and the number of value 1, so m position is arranged to left branch or the right branch of root node.
Preferably, described balanced tree comprises several sub-tree structure, and the organizational form of subtree in storer is: 0th～4 represent the topological structure of this subtree, and it can carry out structural adjustment according to memory bandwidth requirements; 5th～13 indicate the figure place that in subtree topological structure, first node is searched; 59th～67 in order to indicate the figure place of last node representative in subtree topological structure; Whether 68th～75 be leaf node in order to indicate searched next stage; 76th～91 in order to indicate the storer at place; 95th～255 in order to indicate next stage address of node, and successively, every 20 show a node link address.
Compared with prior art, the present invention has following beneficial effect:
1. the storage organization of Trie structure employing stratification in the present invention, reduces the access time of a key to external memory storage, obtains seek rate faster with lower memory cost;
2. in the present invention, adopt balanced tree structure, reduce on the whole the hierarchical structure of tree, search the time thereby reduce.In common binary tree, the time complexity that finds each node is O (n); And adopting balance tree structure of the present invention, the time complexity of searching each node will become O (log
2 n), n represents node number in binary tree;
3. in the present invention, multiple node organizations of this Trie structure are become a stalk tree construction to put in a storage block, this storage data block can be read by an accessing operation.
Brief description of the drawings
By reading the detailed description of non-limiting example being done with reference to the following drawings, it is more obvious that other features, objects and advantages of the present invention will become:
Fig. 1 is the structural drawing that in the present invention, Trie searches;
Fig. 2 is example Trie structural drawing of the present invention;
Fig. 3 is example storage block organizational form of the present invention;
Fig. 4 is the sub-tree structure figure in example of the present invention;
Fig. 5 is the search routine figure of searching structure of the present invention.
Embodiment
Below in conjunction with specific embodiment, the present invention is described in detail.Following examples will contribute to those skilled in the art further to understand the present invention, but not limit in any form the present invention.It should be pointed out that to those skilled in the art, without departing from the inventive concept of the premise, can also make some distortion and improvement.These all belong to protection scope of the present invention.
In the present invention, as shown in Figure 1, the trie searching method being applied in network processing unit provided by the invention, comprises the steps: the structure of Trie search
Step 1: set up multi-level storage organization, described storage organization comprises register file, internal RAM, external SRAM and the outside SDRAM that priority diminishes gradually;
Step 2: build balanced tree, be specially, travel through each of all routing tables, find out the position that the number of its intermediate value 0 and value 1 equates substantially, this is set to current root node, and this position is labeled as and is used, circulate to realize with this foundation that relative equilibrium is set;
Step 3: carry out data storage, wherein, to data storage each time, the stalk tree construction all contiguous several bit data being organized into carries out data storage, and in the process of searching, accessing operation is searched with regard to carrying out long numeric data each time so.For example can carry out four at every turn and search, for the search procedure of 32 routing iinformations in IPv4, even worst case also only need to be read 8 external memory operations, can significantly reduce the memory access time so.
In step 1, adopt multilayered memory structure storage data, as register file, internal RAM, external SRAM and outside SDRAM, its relative velocity and relative capacity as shown in Table 1:
Table one
| ? | Relative velocity | Relative capacity |
| Register file | 100 | 103B |
| Internal RAM | 40 | 105B |
| External SRAM | 5 | 108B |
| Outside SDRAM | 1 | 109B |
Trie structure is in the present invention to reduce the time of memory access, makes the storage mode of following stratification and divides: the root node of Trie is placed in register file; The layer 2-4 of Trie structure is put in internal RAM; The remaining size according to resource is placed in external SRAM or SDRAM.
In the process that key searches each time, the storage of above stratification can be accessed the storer that relative time is larger as far as possible less, thereby reduces access time delay, improves seek rate.
Described step 2 comprises the steps:
Step 2.1: according to all data strip destination data positions, find out the value 0 of n bit data equal with the number of value 1, n position is arranged to the root node of tree;
Step 2.2: differ less according to the value 0 in m position and the number of value 1, so m position is arranged to the left branch of root node; The tree construction of setting up by that analogy as shown in Figure 1.
In the stalk tree shown in Fig. 1, all leave in a storage block of storer, can a memory access all read.
A memory access of DDR2 storer at least can be accessed 256 bit data, and a memory access of DDR3 storer can be accessed 512 bit data, therefore can, according to extract 256 or 512 bit data, carry out long numeric data and search.The long numeric data here is only searched needs several clock period just can complete, and carrys out raising speed by the number of times that reduces memory access.
The organizational form of a kind of sub-tree structure as shown in Figure 3 in storer, is specially 0th～4 topological structures that represent this subtree, carries out structural adjustment according to memory bandwidth requirements; 5th～13 indicate the figure place that in subtree topological structure, first node is searched, by that analogy; 59th～67 in order to indicate the figure place of last node representative in subtree topological structure; Whether 68th～75 be leaf node in order to indicate searched next stage; Which storer is 76th～91 be in order to indicate the storer at place; And 95th～255 in order to indicate next stage address of node, successively, every 20 show a node link address.
In the present embodiment, the Trie search device being applied in network processing unit provided by the invention, comprises as lower device:
Storage organization apparatus for establishing, for setting up multi-level storage organization, described storage organization comprises register file, internal RAM, external SRAM and the outside SDRAM that priority diminishes gradually;
Balanced tree apparatus for establishing, for equilibrium establishment tree, is specially, travel through each of all routing tables, find out the position that the number of its intermediate value 0 and value 1 equates substantially, this is set to current root node, and this position is labeled as and is used, circulate to realize the foundation of relative equilibrium tree with this;
Data storage device, for carrying out data storage, wherein, to data storage each time, is all organized into a stalk tree construction by contiguous several bit data and carries out data storage.
Further, the storage mode of described storage organization is divided into: the root node of Trie is placed in register file; The 2nd to 4 layers of Trie structure are put in internal RAM; The data that Trie structure are greater than to the 4th layer are placed in external SRAM or SDRAM according to the size of resource.
Further, balanced tree apparatus for establishing, first according to all data strip destination data positions, is found out the value 0 of n bit data equal with the number of value 1, n position is arranged to the root node of tree; Secondly differ less according to the value 0 in m position and the number of value 1, so m position is arranged to left branch or the right branch of root node.
Further, described balanced tree comprises several sub-tree structure, and the organizational form of subtree in storer is: 0th～4 represent the topological structure of this subtree, and it can carry out structural adjustment according to memory bandwidth requirements; 5th～13 indicate the figure place that in subtree topological structure, first node is searched; 59th～67 in order to indicate the figure place of last node representative in subtree topological structure; Whether 68th～75 be leaf node in order to indicate searched next stage; 76th～91 in order to indicate the storer at place; 95th～255 in order to indicate next stage address of node, and successively, every 20 show a node link address.
More in preferred embodiment, as shown in Figure 2, suppose total A, a B, C, D, E, F, G, H, s, t, u, v, w, the x, y, z routing tables of totally 16 entry compositions at one, according to finding step of the present invention be:
Step 1: each the 0th of 16 entries of traversal, finds that 0 and 1 number is all 8, so by the 0th root node that is set to this Trie structure.
Step 2: this Trie structure is divided into two groups according to root node, and first group is that tree root branches into 0 node, comprises A, B, C, D, E, F, G, H; Second group is that tree root branches into 1 node, comprises s, t, u, v, w, x, y, z.According to 1,2,3,4,5 equal-order digits, first group is traveled through, can be divided into two groups according to the 2nd again by first group, be labeled as respectively the 3rd group and the 4th group, wherein the 3rd group comprises A, B, C, D, and the 4th group comprises E, F, G, H; Can set up by that analogy a balanced tree, as shown in Figure 2.
Step 3: for this balanced tree, the store status of the 0th of tree root is placed in register file, and other store status is placed in internal RAM, if while setting up a complicated tree, according to circumstances store status is put in external SRAM or outside SDRAM.
Become a subtree all to put in a storage block node organization in a square frame in Fig. 2, this storage block can be read completely by an accessing operation, so the time of searching is in this example (1+n) T, wherein T is the clock period, and n represents once to access the required clock periodicity of internal RAM.And if adopt external memory storage incompressible structure, the memory access time used is 4mT, the required clock periodicity of outside SDRAM storer is once accessed in m representative.General m is 102 left and right, and n is 10 left and right; So can greatly shorten the memory access time in this example.
In the present embodiment, the data structure that the information of tree is mapped to is as shown in Figure 3 stored.Wherein 0th～4 is 0, the institutional framework of pseudo-subtree in representative graph 4, if when interstitial content is not enough to the full sub-tree structure shown in pie graph 4, Gai Shu branch fills structure identical in same layer so that reach balanced tree structure; The value of 5th～13 is 2 to show to search for the first time the second of key, by that analogy; 68th～75 is 1 entirely, shows that this subtree has directly connected leaf node, but not next stage sub-tree structure; In 76th～91, the data of every two are all 01; And 95th～255 to indicate the stored specific address of each nodal information.
Trie search method in network processing unit disclosed by the invention, first set up the binary tree structure of a balance according to the number of each entry intermediate value 0 and value 1, then be placed in a storage block setting adjacent multiple nodes with this Trie, finally this all tree constructions be placed in different memory hierarchys.The benefit that said method brings has: 1, the Trie structure of balance makes the time complexity of searching each node be reduced to O (log
2 n); 2, adopt the pressure texture of stratification to make repeatedly to access long delay and become the storer of access short time delay, thereby greatly shorten the time of memory access, improve seek rate; 3, adopt adjacent multiple node datas are placed in a storage block, can make full use of the high bandwidth of storer, and the number of times of memory access is reduced to original 1/n, obviously reduce the memory access time, significantly improve search efficiency.
The method can make full use of the storage resources of storer, can access the feature of 256 or 512 bit data in conjunction with each memory access, adopts benefit of the present invention to have 2 points:
1, strengthen the utilization factor of memory stores bandwidth.Example as shown in Figure 5, if search and only store an information at every turn, the so each lobus sinister child node that only need to store this node, lobus dexter child node, chooser (whom selection carried out and searched), the information such as key word, their bit wide is determined by the size of look-up table, generally there are six or seven ten, so read and have most wastings of resources in process at every turn, and can be according to the bandwidth of storer in the present invention, determine multiple node storage information to be placed in a storage block, bit wide in an accessing operation will greatly be utilized so, thereby improve the utilization factor of memory bandwidth.
2, improved seek rate.General binary tree accessing operation is equivalent to carry out one to be searched, and when reading so, just need to carry out four times accessing operation for four; And when these four being searched to node and are placed in same storage block in employing the present invention, in to these four processes of searching, only need to access primary memory operation so, significantly reduce Memory accessing delay.
Above specific embodiments of the invention are described.It will be appreciated that, the present invention is not limited to above-mentioned specific implementations, and those skilled in the art can make various distortion or amendment within the scope of the claims, and this does not affect flesh and blood of the present invention.
Claims (8)
1. be applied to the Trie search method in network processing unit, it is characterized in that, comprise the steps:
Step 1: set up multi-level storage organization, described storage organization comprises register file, internal RAM, external SRAM and the outside SDRAM that priority diminishes gradually;
Step 2: equilibrium establishment tree, be specially, travel through each of all routing tables, find out the immediate position of number of its intermediate value 0 and value 1, this is set to current root node, and this position is labeled as and is used, circulate to realize the foundation of relative equilibrium tree with this;
Step 3: carry out data storage, wherein, to data storage each time, the sub-tree structure all contiguous several bit data being organized into is carried out data storage.
2. the Trie search method being applied in network processing unit according to claim 1, is characterized in that,
The storage mode of described storage organization is divided into: the root node of Trie is placed in register file; The 2nd to 4 layers of Trie structure are put in internal RAM; The data that Trie structure are greater than to the 4th layer are placed in external SRAM or SDRAM according to the size of resource.
3. the Trie search method being applied in network processing unit according to claim 1 and 2, is characterized in that, described step 2 comprises the steps:
Step 2.1: according to all data strip destination data positions, if the number of the value of n bit data 0 and value 1 is equal, n position is arranged to the root node of tree;
Step 2.2: differ less according to the value 0 in m position and the number of value 1, so m position is arranged to left branch or the right branch of root node.
4. the Trie search method being applied in network processing unit according to claim 1 and 2, it is characterized in that, described balanced tree comprises several sub-tree structure, the organizational form of subtree in storer is: 0th～4 represent the topological structure of this subtree, and it can carry out structural adjustment according to memory bandwidth requirements; 5th～13 indicate the figure place that in subtree topological structure, first node is searched; 59th～67 in order to indicate the figure place of last node representative in subtree topological structure; Whether 68th～75 be leaf node in order to indicate searched next stage; 76th～91 in order to indicate the storer at place; 95th～255 in order to indicate next stage address of node, and successively, every 20 show a node link address.
5. be applied to the Trie search device in network processing unit, it is characterized in that, comprise as lower device:
Storage organization apparatus for establishing, for setting up multi-level storage organization, described storage organization comprises register file, internal RAM, external SRAM and the outside SDRAM that priority diminishes gradually;
Balanced tree apparatus for establishing, for equilibrium establishment tree, is specially, travel through each of all routing tables, find out immediate of the number of its intermediate value 0 and value 1, this is set to current root node, and this position is labeled as and is used, circulate to realize the foundation of relative equilibrium tree with this;
Data storage device, for carrying out data storage, wherein, to data storage each time, the sub-tree structure all contiguous several bit data being organized into is carried out data storage.
6. the Trie search device being applied in network processing unit according to claim 5, is characterized in that,
The storage mode of described storage organization is divided into: the root node of Trie is placed in register file; The 2nd to 4 layers of Trie structure are put in internal RAM; The data that Trie structure are greater than to the 4th layer are placed in external SRAM or SDRAM according to the size of resource.
7. according to the Trie search device in network processing unit that is applied to described in claim 5 or 6, it is characterized in that, balanced tree apparatus for establishing, first according to all data strip destination data positions, is found out the value 0 of n bit data equal with the number of value 1, n position is arranged to the root node of tree; Secondly differ less according to the value 0 in m position and the number of value 1, so m position is arranged to left branch or the right branch of root node.
8. according to the Trie search device in network processing unit that is applied to described in claim 5 or 6, it is characterized in that, described balanced tree comprises several sub-tree structure, the organizational form of subtree in storer is: 0th～4 represent the topological structure of this subtree, and it can carry out structural adjustment according to memory bandwidth requirements; 5th～13 indicate the figure place that in subtree topological structure, first node is searched; 59th～67 in order to indicate the figure place of last node representative in subtree topological structure; Whether 68th～75 be leaf node in order to indicate searched next stage; 76th～91 in order to indicate the storer at place; 95th～255 in order to indicate next stage address of node, and successively, every 20 show a node link address.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410307241.0A CN104090942A (en) | 2014-06-30 | 2014-06-30 | Trie search method and device applied to network processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410307241.0A CN104090942A (en) | 2014-06-30 | 2014-06-30 | Trie search method and device applied to network processor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104090942A true CN104090942A (en) | 2014-10-08 |
Family
ID=51638658
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410307241.0A Pending CN104090942A (en) | 2014-06-30 | 2014-06-30 | Trie search method and device applied to network processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104090942A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335300A (en) * | 2014-12-10 | 2016-02-17 | 成都科来软件有限公司 | Data storage method and device |
| CN108733678A (en) * | 2017-04-14 | 2018-11-02 | 华为技术有限公司 | A kind of method, apparatus and relevant device of data search |
| WO2018205151A1 (en) * | 2017-05-09 | 2018-11-15 | 华为技术有限公司 | Data updating method and storage device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040008634A1 (en) * | 2002-05-31 | 2004-01-15 | Vijay Rangarajan | Method and apparatus for generating and using enhanced tree bitmap data structures in determining a longest prefix match |
| CN1761238A (en) * | 2004-09-14 | 2006-04-19 | 阿尔卡特公司 | Optimization of routing forwarding database in a network processor |
| CN103051543A (en) * | 2012-11-01 | 2013-04-17 | 广州微仕科信息技术有限公司 | Route prefix processing, lookup, adding and deleting method |
| CN103404092A (en) * | 2011-11-30 | 2013-11-20 | 华为技术有限公司 | Route prefix storage method and device and route address searching method and device |
-
2014
- 2014-06-30 CN CN201410307241.0A patent/CN104090942A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040008634A1 (en) * | 2002-05-31 | 2004-01-15 | Vijay Rangarajan | Method and apparatus for generating and using enhanced tree bitmap data structures in determining a longest prefix match |
| CN1761238A (en) * | 2004-09-14 | 2006-04-19 | 阿尔卡特公司 | Optimization of routing forwarding database in a network processor |
| CN103404092A (en) * | 2011-11-30 | 2013-11-20 | 华为技术有限公司 | Route prefix storage method and device and route address searching method and device |
| CN103051543A (en) * | 2012-11-01 | 2013-04-17 | 广州微仕科信息技术有限公司 | Route prefix processing, lookup, adding and deleting method |
Non-Patent Citations (1)
| Title |
|---|
| 张琦 等: ""Trie 树路由查找算法在网络处理器中的实现"", 《计算机工程》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335300A (en) * | 2014-12-10 | 2016-02-17 | 成都科来软件有限公司 | Data storage method and device |
| CN105335300B (en) * | 2014-12-10 | 2018-10-12 | 成都科来软件有限公司 | A kind of date storage method and device |
| CN108733678A (en) * | 2017-04-14 | 2018-11-02 | 华为技术有限公司 | A kind of method, apparatus and relevant device of data search |
| CN108733678B (en) * | 2017-04-14 | 2021-11-09 | 华为技术有限公司 | Data searching method, device and related equipment |
| WO2018205151A1 (en) * | 2017-05-09 | 2018-11-15 | 华为技术有限公司 | Data updating method and storage device |
| CN110168532A (en) * | 2017-05-09 | 2019-08-23 | 华为技术有限公司 | Data-updating method and storage device |
| CN110168532B (en) * | 2017-05-09 | 2021-08-20 | 华为技术有限公司 | Data updating method and storage device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9269411B2 (en) | Organizing data in a hybrid memory for search operations | |
| CN102484610B (en) | Routing table construction method and device and routing table lookup method and device | |
| US9077669B2 (en) | Efficient lookup methods for ternary content addressable memory and associated devices and systems | |
| Le et al. | Scalable tree-based architectures for IPv4/v6 lookup using prefix partitioning | |
| CN103051543B (en) | A kind of process of route prefix, search, increase and delet method | |
| CN102035727B (en) | Method and equipment for processing routing | |
| CN107153707A (en) | A kind of Hash table construction method and system for nonvolatile memory | |
| CN106416152A (en) | Lookup device, lookup configuration method and lookup method | |
| CN103745008A (en) | Sorting method for big data indexing | |
| CN101621502A (en) | Method and device for storing and searching routing table | |
| CN108134739B (en) | Route searching method and device based on index trie | |
| CN104158744A (en) | Method for building table and searching for network processor | |
| CN102662855B (en) | Storage method and system of binary tree | |
| CN103914483A (en) | File storage method and device and file reading method and device | |
| CN107330094A (en) | The Bloom Filter tree construction and key-value pair storage method of dynamic memory key-value pair | |
| CN104090942A (en) | Trie search method and device applied to network processor | |
| CN101277252A (en) | Method for traversing multi-branch Trie tree | |
| CN106708749A (en) | Data search method | |
| US10176060B2 (en) | Memory apparatus for applying fault repair based on physical region and virtual region and control method thereof | |
| US9396286B2 (en) | Lookup with key sequence skip for radix trees | |
| CN110460528A (en) | Name the FIB storage organization and its application method of data network Forwarding plane | |
| US9361404B2 (en) | Offline radix tree compression with key sequence skip | |
| CN103546167B (en) | Code translator and the method that parsing data are decoded | |
| CN114579537A (en) | Distributed graph database optimization method and device, electronic equipment and storage medium | |
| Erdem et al. | Value-coded trie structure for high-performance IPv6 lookup |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141008 |
|
| RJ01 | Rejection of invention patent application after publication |