WO2001067456A2 - Multiple entry matching in a content addressable memory - Google Patents
Multiple entry matching in a content addressable memory Download PDFInfo
- Publication number
- WO2001067456A2 WO2001067456A2 PCT/SE2001/000483 SE0100483W WO0167456A2 WO 2001067456 A2 WO2001067456 A2 WO 2001067456A2 SE 0100483 W SE0100483 W SE 0100483W WO 0167456 A2 WO0167456 A2 WO 0167456A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- carry
- data
- comparand
- signal
- compare
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C15/00—Digital stores in which information comprising one or more characteristic parts is written into the store and in which information is read-out by searching for one or more of these characteristic parts, i.e. associative or content-addressed stores
- G11C15/04—Digital stores in which information comprising one or more characteristic parts is written into the store and in which information is read-out by searching for one or more of these characteristic parts, i.e. associative or content-addressed stores using semiconductor elements
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C15/00—Digital stores in which information comprising one or more characteristic parts is written into the store and in which information is read-out by searching for one or more of these characteristic parts, i.e. associative or content-addressed stores
Definitions
- the present invention relates to computer memory devices, and more particularly to content addressable memories.
- a conventional Content Addressable Memory (CAM) 101 is illustrated in FIG. 1.
- a CAM 101 can be viewed as a computer accessible storage device having an inverse type of access compared with typical addressable memories.
- a typical Random Access Memory (RAM) is accessed by means of an address port. One supplies an address to the address port, and the RAM
- RAM supplies at its output the data value that is stored at the memory location designated by the address.
- one accesses a CAM 101 by supplying data of interest (referred to as a "comparand") at a comparand input port 103.
- the comparand is represented as a ternary value for each bit: "0", “ 1 " or "*", the last one interpreted as a wild card.
- Each storage entry 105 in the CAM 101 includes, or is otherwise associated with, logic that compares the value stored in the entry 105 with the comparand.
- the resulting signals are supplied to a priority encoder 107, which generates an address 109 that indicates one of the matching entries (e.g., the lowest address associated with a matching entry 105).
- IP CAMs are increasingly used to implement massively parallel searches over a large number of data.
- IP Internet Protocol
- the forwarding and classification of packets each require a search of a large to massive data base.
- IP forwarding lookup is the problem of using the destination address of an incoming packet to find a matching one from among tens of thousands of routing table entries. Once a best matching entry is found, it is used to determine the next hop for the packet.
- the routing table may contain several matching entries, where the one with the least number of "wild cards" is considered the best match.
- IP classification is the more general problem of using selected parts of the IP and higher level headers of an incoming packet to find a matching one from among a hundred to a thousand classification entries. Once a highest priority entry is found, it is used for different purposes, such as filtering, IP security selection and virtual routing.
- the IP forwarding lookup problem can be addressed by extending the basic CAM principle to allow the bits of stored data entries to take on wild card values.
- CAMs extended in this manner are called ternary CAMs. With such an extended CAM, the routing table entries are stored directly in the CAM, one after another. The destination address of the packet is simply presented as a comparand (without wild cards) to the CAM, which performs the matching in a single access. The address generated by the CAM is used to read further (non- associative) routing data from a standard RAM.
- the task of updating the CAM is simple, since each data directly corresponds to a routing entry.
- entries should be stored in order based on the number of wild cards. Hence, relocation can be necessary when updates are made.
- the IP classification problem is improved, but not fully addressed by conventional CAMs. Considering all possible matching strategies, for the moment, it is apparent that the trivial method of matching against every table entry by reading them one after another will not achieve performance requirements. Hence, several methods using tree representations have been used. More particularly, the routing table may be compiled into a tree, which is stored in a
- the tree is traversed from the root to the leaves. In every node, a subfield of the destination address is used together with node specific data to decide what branch of the tree to select as the next step.
- the routing entries are associated with some nodes or leaves in the tree. Eventually, a best match will be found.
- CAMs does not provide a straightforward solution to combining the results in the above-described second step of the multidimensional matching problem.
- a conventional CAM is arranged as number of entries, each holding a data word of a fixed width. Matching is limited to this width.
- Conventional CAMs offer no general and fast way to make matches to a comparand greater than the width of a cell.
- a content addressable memory that comprises a matrix of cells having (n + 1) columns and m rows, wherein n and m are each integers greater than or equal to 1 , and wherein each row of cells comprises: n data storage cells, a carry storage cell, word compare logic, and carry write logic.
- Each of the n data storage cells comprises: storage logic for storing a respective one of n data signals; an input for receiving a respective one of n comparand data signals; and bit compare logic that generates a bit compare signal that indicates whether the stored one of the n data signals matches the received one of the n comparand data signals.
- the carry storage cell comprises: a first input for receiving a carry data signal; storage logic for storing the carry data signal; a second input for receiving a carry comparand data signal; and carry compare logic that generates a carry compare signal that indicates whether the stored carry data signal matches the received carry comparand data signal.
- the word compare logic generates a word compare signal that indicates whether each of the bit compare signals in the row indicates that the stored one of the n data signals matches the received one of the n comparand data signals and that the carry compare signal in the row indicates that the stored carry data signal matches the received carry comparand data signal.
- the carry write logic generates a carry storage cell write signal only if the word compare signal generated by the word compare logic in a previous row of cells indicates that each of the bit compare signals in the previous row indicates that the stored one of the n data signals matches the received one of the n comparand data signals and that the carry compare signal in the previous row indicates that the stored carry data signal matches the received carry comparand data signal.
- each of the n comparand data signals and the carry comparand data signals may be a ternary value selected from a group of values consisting of "0", "1" and “don't care”; and the "don't care" value matches any of the ternary values.
- the carry write logic further comprises a write input port for receiving a carry write signal, and the carry write logic generates the carry storage cell write signal only if the carry write signal is asserted and the word compare signal generated by the word compare logic in a previous row of cells indicates that each of the bit compare signals in the previous row indicates that the stored one of the n data signals matches the received one of the n comparand data signals and that the carry compare signal in the previous row indicates that the stored carry data signal matches the received carry comparand data signal.
- the word compare logic comprises the bit compare signals in the row being "wire ANDed" together.
- the operating techniques include: storing one or more data items in the content addressable memory, wherein each data item comprises a plurality of fractional parts that are distinguishable from one another, and wherein for each data item, the fractional parts are stored in different rows of the content addressable memory; clearing the stored carry data signal in each of the rows; and sequentially supplying fractional parts of a comparand data item to the inputs to the n data storage cells and to the second inputs of the carry storage cells, wherein a first one of the fractional parts of the comparand data item comprises a carry part that is not set equal to a " 1 " value; and remaining ones of the fractional parts of the comparand data item each comprise a carry part that is set equal to a " 1 " value, whereby assertion of one of the word compare signals indicates that all of the fractional parts of a stored data item matched the respective supplied fractional parts of the comparand data item.
- each fractional part of the comparand data item may comprise: a comparand data fraction part; a comparand tag part; and a carry part.
- the comparand tag part may be useful, for example, for specifying what position the fractional part occupies in the sequence of fractional parts that make up the comparand data item.
- FIG. 1 is a block diagram of a conventional content addressable memory
- FIG. 2 is a block diagram of a memory matrix of a conventional
- FIG. 3 is a block diagram of a memory matrix that has been extended in accordance with an aspect of the invention.
- FIG. 4 is an exemplary technique for using the inventive CAM to perform a matching operation over multiple entries.
- the invention relates to methods and devices for matching over multiple entries in a CAM. In accordance with an aspect of the invention, this is achieved by providing special carry-cell logic in the memory.
- the carry-cell logic permits each cell to generate a match based not only on whether that cell's contents match a presently applied comparand, but also whether a neighboring cell matched a previously applied comparand.
- Computer readable storage media include, but are not limited to, static or dynamic Random Access Memory (SRAM or DRAM), Read Only Memory (ROM), magnetic storage media (e.g., disk, diskettes or tape) and optical media (e.g., Compact Disk ROM, or CD ROM).
- SRAM static or dynamic Random Access Memory
- ROM Read Only Memory
- magnetic storage media e.g., disk, diskettes or tape
- optical media e.g., Compact Disk ROM, or CD ROM
- FIG. 2 this is a block diagram of a memory matrix
- the CAM 101 is organized as m entries, each holding a word of n bits.
- the n vertical data lines hold the comparand input data during a match.
- the m horizontal "wired AND" lines, & j compute the result of a match.
- This result is amplified by amplifiers 203 and presented as an output to the priority encoder (not shown in FIG. 2) on m matching signals ⁇ .
- the logic for writing data into the cells is well known and not relevant to the invention. Thus, for the sake of simplicity, the data writing logic has been omitted from the figure.
- both the comparand data and the stored data are ternary values ⁇ 0, 1,* ⁇ , where "*" represents a "don't care” state that matches anything.
- the n horizontal data lines are usually implemented as double physical lines where the ternary values are coded as a pair of binary values. The actual coding is not important to the invention, however, so in this description the abstraction of single data lines carrying ternary values is maintained.
- FIG. 3 is a block diagram of a similar memory matrix 301 that has been extended in accordance with an aspect of the invention.
- a special CAM carry cell V j is introduced for every word i.
- Each of the CAM carry cells V j stores a carry data bit, C j , that mirrors the result of a previous match result for the previous word i-1 (e.g., in the exemplary embodiment the previous word is the word immediately above the present word under consideration).
- the CAM carry cell V j participates as an ordinary cell during a matching operation.
- word compare logic &J is provided that generates a word compare signal m j that indicates whether each of the bit compare signals in the row indicates that the stored one of the n data signals S j : matches the received one of the n comparand data signals d: and that the carry compare signal in the row indicates that the stored carry data signal C j matches the received carry comparand data signal d 0
- the CAM carry cell V j may also be written as a result of the matching operation.
- a vertical carry data line, cd holds the value to be optionally written, and a vertical carry write line, cw, must be asserted in order for the CAM carry cells V j to be conditionally written.
- the phrase "conditionally written" has been used in the preceding sentence because the cw line does not, by itself, determine whether the value on the cd line will be clocked into a CAM carry cell V j .
- carry write logic 303 that generates a carry write signal cW j that enables the carry write operation to be performed if and only if the carry write line, cw, is active and there is a match for the immediately preceding word, i-1 (the latter condition being indicated by the assertion of the matching signal, m ⁇ ).
- the carry write logic 303 is implemented as a logical AND gate. However, in alternative embodiments, equivalent logic configurations could instead be used.
- rri Q has been introduced as a new input to the cell matrix.
- the value of I ⁇ I Q is 0 unless otherwise noted.
- one more horizontal data line, d 0 has been introduced as a new input to the cell matrix for carrying the comparand value that is supplied to each of the CAM carry cells V j .
- the exemplary device described above with respect to FIG. 3 can be used for matching over multiple entries.
- An exemplary technique for doing this will now be described with reference to FIG. 4.
- a stored data item occupying more than a single entry is partitioned into q smaller fractions denoted by S 1 to S q .
- a unique tag value, t, ranging from 1 to q is introduced for each type of fraction.
- Each entry in the CAM holds a data fraction appended by its corresponding tag value. Fractions belonging to the same original data are stored in successive entries (e.g. , immediately following one another) and in the internal order of their corresponding tag values. The fractions and their corresponding tags are stored in the conventional CAM cells, U j n ...u j j . A sufficient number of the conventional CAM cells should be allocated to the tag bits to permit representing the highest tag value, q. The remaining conventional CAM cells for each word may be allocated to storing one of the fractions.
- the CAM carry cell, V j initially stores an unknown value, depicted in FIG. 4 by the "?" symbol.
- the comparand value is similarly partitioned into corresponding fractions, denoted herein by D 1 to D .
- D j to D some of the bits represent fractional data to be matched, other bits represent tag bits to be matched, and another bit represents a carry bit to be matched.
- the purpose of the tag bits is to enable the marking of a fractional part as being associated with a particular position (e.g. , first, second, third, ...) within the complete stored data item, so that when a matching operation is to be performed, a comparand value can specify not only the desired fractional data part, but also the tag value to indicate which fractional part is being searched.
- the algorithm for matching over the multiple fractions may be specified as follows (where text to the right of the "/*" indicates a comment):
- the comparand data is "don't care", which means that every entry matches. As a result, all carry cells will be initialized with a 0 value.
- cycle 1 the first fraction of the comparand appended by a tag value of 1 and a "don't care" value on the carry part is presented on the comparand data lines. All entries that have a first data fraction matching the first comparand data will match. The result is that for each such entry, the carry cell of the next entry, holding the second data fraction, will be written with a " 1" value.
- the first fraction of the comparand appended by a tag value of 2 and a carry value of 1 is presented on the comparand data lines.
- a priority encoder such as the priority encoder 107 shown in FIG. 1 will present the address of the last fraction of a complete stored data that matches a complete operand.
- the invention is useful for storing and matching against arbitrarily long data in a CAM. For some special cases, other specific methods using repeated accesses to the CAM can be used to match long data. However, the latency is much higher and the memory consumption is often higher.
- the invention can be used in all technical areas containing a time critical non-trivial matching problem.
- One example is the IP classification problem present in Internet routers, as described in the BACKGROUND section.
- To apply the inventive CAM to solve such a problem different relevant fields of the classification entries are stored at consecutive addresses of the CAM.
- the inventive CAM allows logical "AND" conditions to be performed between consecutive entries over several cycles, hence allowing for matching on arbitrarily long data. The result is immediately available as an address after the last cycle without any combination of results needed.
- data fractions other than the first and the last one, can be omitted where all bits have "don't care" as a value.
- the tag values do not need to be the ones indicated above in the exemplary embodiment. Instead, any values will do as long as the order between different types of fractions is kept.
- the comparand tag part may be set to any value that distinguishes the fractional part from other fractional parts of the comparand data item. In the matching algorithm, this means that the constants 1 to q used for the tag can be replaced by distinguishable variables T ⁇ to T
- the structure of the inventive CAM permits the classifications to be performed more efficiently, by presenting the parameters (together with their respective tags) as comparand data items to the CAM in the following order:
- the exemplary embodiment illustrates data being stored into consecutive rows in a top-to-bottom direction.
- alternative embodiments of the invention can just as easily be arranged to function in other directions, such as from bottom-to-top. Consequently, throughout this description as well as in the claims, the use of terms such as “previous”, “next” and the like shall not be construed to refer only to the directions illustrated in the exemplary embodiment, but shall instead more generally refer to that which would be "previous” or "next (etc.) as determined relative to a predetermined orientation, regardless of whether that predetermined orientation is top-to-bottom, bottom-to- top, or other.
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2001239616A AU2001239616A1 (en) | 2000-03-08 | 2001-03-07 | Multiple entry matching in a content addressable memory |
EP01914269A EP1281178B1 (en) | 2000-03-08 | 2001-03-07 | Multiple entry matching in a content addressable memory |
DE60102230T DE60102230T2 (en) | 2000-03-08 | 2001-03-07 | MULTIPLE ENTRY COMPARISON IN A CONTINUOUS ADDRESSABLE MEMORY |
AT01914269T ATE261177T1 (en) | 2000-03-08 | 2001-03-07 | MULTIPLE ENTRY COMPARISON IN A CONTENT ADDRESSABLE STORAGE |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US18789500P | 2000-03-08 | 2000-03-08 | |
US60/187,895 | 2000-03-08 | ||
US09/666,844 US6259620B1 (en) | 2000-03-08 | 2000-09-21 | Multiple entry matching in a content addressable memory |
US09/666,844 | 2000-09-21 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2001067456A2 true WO2001067456A2 (en) | 2001-09-13 |
WO2001067456A3 WO2001067456A3 (en) | 2002-02-07 |
Family
ID=26883517
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/SE2001/000483 WO2001067456A2 (en) | 2000-03-08 | 2001-03-07 | Multiple entry matching in a content addressable memory |
Country Status (6)
Country | Link |
---|---|
US (1) | US6259620B1 (en) |
EP (1) | EP1281178B1 (en) |
AT (1) | ATE261177T1 (en) |
AU (1) | AU2001239616A1 (en) |
DE (1) | DE60102230T2 (en) |
WO (1) | WO2001067456A2 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004044920A3 (en) * | 2002-11-13 | 2004-07-29 | Mosaid Technologies Inc | Method and apparatus for wide word deletion in content addressable memories |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6615210B1 (en) * | 2000-02-04 | 2003-09-02 | Broad Web Corporation | Bit stream ternary match scheme |
JP2002237190A (en) * | 2001-02-07 | 2002-08-23 | Kawasaki Microelectronics Kk | Associative memory device and its constituting method |
US6512684B2 (en) * | 2001-06-11 | 2003-01-28 | International Business Machines Corporation | Content addressable memory having cascaded sub-entry architecture |
US7181568B2 (en) * | 2004-03-25 | 2007-02-20 | Intel Corporation | Content addressable memory to identify subtag matches |
US20060212426A1 (en) * | 2004-12-21 | 2006-09-21 | Udaya Shakara | Efficient CAM-based techniques to perform string searches in packet payloads |
US20060190679A1 (en) * | 2005-02-18 | 2006-08-24 | Albrecht Alan R | Content addressable memory supporting multiple width searches |
US8050185B2 (en) * | 2005-08-24 | 2011-11-01 | Hewlett-Packard Development Company, L.P. | Sampling of network traffic based on CAM lookup |
RU2509383C2 (en) * | 2012-04-06 | 2014-03-10 | Федеральное государственное бюджетное образовательное учреждение высшего профессионального образования "Юго-Западный государственный университет" (ЮЗГУ) | Method for parallel search and row replacement and homogeneous memory matrix for realising said method |
US9859006B1 (en) * | 2016-06-17 | 2018-01-02 | Globalfoundries Inc. | Algorithmic N search/M write ternary content addressable memory (TCAM) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4959811A (en) * | 1986-11-03 | 1990-09-25 | Texas Instruments Incorporated | Content addressable memory including comparison inhibit and shift register circuits |
US5440715A (en) * | 1990-06-27 | 1995-08-08 | Advanced Micro Devices, Inc. | Method and apparatus for expanding the width of a content addressable memory using a continuation bit |
US6000008A (en) * | 1993-03-11 | 1999-12-07 | Cabletron Systems, Inc. | Method and apparatus for matching data items of variable length in a content addressable memory |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5226005A (en) | 1990-11-19 | 1993-07-06 | Unisys Corporation | Dual ported content addressable memory cell and array |
JP2728819B2 (en) | 1991-12-18 | 1998-03-18 | 川崎製鉄株式会社 | Semiconductor integrated circuit |
JP2775549B2 (en) | 1992-05-08 | 1998-07-16 | 三菱電機株式会社 | Associative memory cell and associative memory circuit |
GB9213821D0 (en) | 1992-06-30 | 1992-08-12 | Inmos Ltd | Content addressable memory |
US5568415A (en) | 1993-02-19 | 1996-10-22 | Digital Equipment Corporation | Content addressable memory having a pair of memory cells storing don't care states for address translation |
US5454094A (en) | 1993-06-24 | 1995-09-26 | Hal Computer Systems, Inc. | Method and apparatus for detecting multiple matches in a content addressable memory |
JP3169155B2 (en) | 1993-12-22 | 2001-05-21 | インターナショナル・ビジネス・マシーンズ・コーポレ−ション | Circuit for caching information |
US5568416A (en) | 1994-03-24 | 1996-10-22 | Kawasaki Steel Corporation | Associative memory |
US5446686A (en) | 1994-08-02 | 1995-08-29 | Sun Microsystems, Inc. | Method and appartus for detecting multiple address matches in a content addressable memory |
JP2812262B2 (en) | 1995-08-31 | 1998-10-22 | 日本電気株式会社 | Associative memory |
US5796758A (en) | 1996-10-08 | 1998-08-18 | International Business Machines Corporation | Self-checking content-addressable memory and method of operation for detecting multiple selected word lines |
US6081440A (en) * | 1998-11-05 | 2000-06-27 | Lara Technology, Inc. | Ternary content addressable memory (CAM) having fast insertion and deletion of data values |
US5999435A (en) | 1999-01-15 | 1999-12-07 | Fast-Chip, Inc. | Content addressable memory device |
US6108227A (en) * | 1999-07-23 | 2000-08-22 | Lara Technology, Inc. | Content addressable memory having binary and ternary modes of operation |
-
2000
- 2000-09-21 US US09/666,844 patent/US6259620B1/en not_active Expired - Lifetime
-
2001
- 2001-03-07 WO PCT/SE2001/000483 patent/WO2001067456A2/en active IP Right Grant
- 2001-03-07 EP EP01914269A patent/EP1281178B1/en not_active Expired - Lifetime
- 2001-03-07 AU AU2001239616A patent/AU2001239616A1/en not_active Abandoned
- 2001-03-07 DE DE60102230T patent/DE60102230T2/en not_active Expired - Lifetime
- 2001-03-07 AT AT01914269T patent/ATE261177T1/en not_active IP Right Cessation
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4959811A (en) * | 1986-11-03 | 1990-09-25 | Texas Instruments Incorporated | Content addressable memory including comparison inhibit and shift register circuits |
US5440715A (en) * | 1990-06-27 | 1995-08-08 | Advanced Micro Devices, Inc. | Method and apparatus for expanding the width of a content addressable memory using a continuation bit |
US6000008A (en) * | 1993-03-11 | 1999-12-07 | Cabletron Systems, Inc. | Method and apparatus for matching data items of variable length in a content addressable memory |
Non-Patent Citations (1)
Title |
---|
R P MAYER: "A variable access associative memory chip for flexible searches in RAM" CONFERENCE PROCEEDINGS OF THE 1995 IEEE FOURTEENTH ANNUAL INTERNATIONAL PHOENIX CONFERENCE ON COMPUTERS AND COMMUNICATIONS, 1995, pages 102-108, XP002901888 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004044920A3 (en) * | 2002-11-13 | 2004-07-29 | Mosaid Technologies Inc | Method and apparatus for wide word deletion in content addressable memories |
US7136961B2 (en) | 2002-11-13 | 2006-11-14 | Mosaid Technologies, Inc. | Method and apparatus for wide word deletion in content addressable memories |
US7558909B2 (en) | 2002-11-13 | 2009-07-07 | Satech Group A.B. Limited Liability Company | Method and apparatus for wide word deletion in content addressable memories |
CN1739164B (en) * | 2002-11-13 | 2011-02-16 | 萨特克集团有限责任公司 | Method and apparatus for wide word deletion in content addressable memories |
Also Published As
Publication number | Publication date |
---|---|
ATE261177T1 (en) | 2004-03-15 |
WO2001067456A3 (en) | 2002-02-07 |
US6259620B1 (en) | 2001-07-10 |
DE60102230T2 (en) | 2005-03-03 |
DE60102230D1 (en) | 2004-04-08 |
AU2001239616A1 (en) | 2001-09-17 |
EP1281178A2 (en) | 2003-02-05 |
EP1281178B1 (en) | 2004-03-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6910097B1 (en) | Classless interdomain routing using binary content addressable memory | |
US6526474B1 (en) | Content addressable memory (CAM) with accesses to multiple CAM arrays used to generate result for various matching sizes | |
KR101714558B1 (en) | Using storage cells to perform computation | |
US6839800B2 (en) | RAM-based range content addressable memory | |
Kirsch et al. | Hash-based techniques for high-speed packet processing | |
US6490279B1 (en) | Fast data base research and learning apparatus | |
US7565482B1 (en) | Method and device for scalable multiple match extraction from search data | |
US20040139274A1 (en) | Virtual content addressable memory with high speed key insertion and deletion and pipelined key search | |
US6259620B1 (en) | Multiple entry matching in a content addressable memory | |
US20110283061A1 (en) | Updating cam arrays using prefix length distribution prediction | |
EP1678619B1 (en) | Associative memory with entry groups and skip operations | |
JP3599273B2 (en) | Improvement of memory that can refer to contents | |
US10297291B2 (en) | Semiconductor device | |
US7219188B1 (en) | Segmented content addressable memory array and priority encoder | |
US20060190679A1 (en) | Content addressable memory supporting multiple width searches | |
US6421660B1 (en) | Enhanced searching method and apparatus for variable bit chains | |
US7228378B1 (en) | Entry location in a content addressable memory | |
JP3691018B2 (en) | Longest match search circuit and method, program, and recording medium | |
US20040128440A1 (en) | Distributed programmable priority encoder capable of finding the longest match in a single operation | |
US11886746B1 (en) | Algorithmic TCAM with storage activity-based read | |
US9870173B2 (en) | Apparatus and method for optimized n-write/1-read port memory design | |
US6715028B1 (en) | Data selecting memory device and selected data transfer device | |
JP6170718B2 (en) | Search system | |
KR970008197A (en) | Sequential Multiple Search / Comparison Method and Apparatus | |
JPS5994147A (en) | Stack control system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A2 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT TZ UA UG UZ VN YU ZA ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A2 Designated state(s): GH GM KE LS MW MZ SD SL SZ TZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
AK | Designated states |
Kind code of ref document: A3 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT TZ UA UG UZ VN YU ZA ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A3 Designated state(s): GH GM KE LS MW MZ SD SL SZ TZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2001914269 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2001914269 Country of ref document: EP |
|
WWG | Wipo information: grant in national office |
Ref document number: 2001914269 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: JP |