64 results about "Match line" patented technology

In an implementation, a device includes: a first inverter; a first sense amplifier connected to the first inverter; a first match line connected to the first sense amplifier; a first analog content addressable memory cell including: a first pull-down transistor; a second pull-down transistor, the first pull-down transistor and the second pull-down transistor being connected in series between the first match line and a reference node; a first memory circuit configured to store a first upper bound of a first range and to activate the first pull-down transistor when an analog input value is less than the first upper bound of the first range; and a second memory circuit configured to store a first lower bound of the first range and to activate the second pull-down transistor when the analog input value is greater than the first lower bound of the first range.

Examples of the presently disclosed technology provide CAM-based circuits specially constructed to implement Boolean satisfiability problems involving k-XOR-SAT clauses. With the strategic addition of auxiliary counting and logic circuits that evaluate match line voltage outputs of a CAM at k discrete times in order to determine whether a counted number of matches returned by a match line satisfies a pre-determine parity condition—where k represents a number of literals of a k-XOR-SAT clause of a Boolean satisfiability problem—a circuit of the present technology can leverage a common CAM (i.e., the same CAM) to implement the k-XOR-SAT clause and k-SAT clauses. Accordingly, this extremely versatile circuit can be used to implement k-XOR-SAT and k-SAT-k-XOR-SAT hybrid problems in less time, and with less hardware and power consumption than existing hardware accelerators.

The invention relates to the technical field of computing resource scheduling, in particular to a GPU server resource dynamic allocation management method, device and equipment and a medium, the GPU server resource dynamic allocation management method comprises the steps of extracting a jump path to divide task migration segments, matching path repeated tasks in a thread, screening overlapped tasks in combination with scheduling and residence time, and performing dynamic allocation management on GPU server resources. And analyzing a resource access trend to identify scrambling stable combinations, and associating address frequencies to generate a dynamic allocation list. According to the method, the behavior similarity among multiple tasks is identified by extracting the jump path among the tasks and performing repetition analysis, the dense overlapping interval of task execution is accurately delimited by means of cross mapping of task scheduling time and path residence time, and the task performance of the multiple tasks is determined in combination with the derivative judgment mode of the change trend of the access frequency of the shared resources. And identifying a task combination with stable resource scrambling behavior distribution, and establishing a dynamic binding strategy between the task and the GPU address resource based on the joint sorting of the address access frequency and the residence duration.

A content-addressable memory includes a pre-search content-addressable memory array, a control circuit, and a main search content-addressable memory array. The pre-search content-addressable memory array is configured to perform a first data search to selectively adjust levels of a plurality of match lines in the pre-search content-addressable memory array. The control circuit is configured to generate a detection signal according to the levels of the plurality of match lines after the first data search is performed, and to generate an enable signal according to the detection signal. The main search content-addressable memory array is configured to selectively perform a second data search according to the enable signal.

A content addressable memory based on a self-rectifying ferroelectric tunnel junction element comprises: a cell array unit having a plurality of TCAM cells, each comprising two self-rectifying ferroelectric tunnel junction elements (SR-FTJ) connected between a corresponding match line of a plurality of match lines extending in a first direction and a corresponding bit line pair of a plurality of bit line pairs extending in a second direction; a precharge unit precharging a corresponding match line of the plurality of match lines to a power supply voltage level in response to a precharge signal; and a data input / output unit having a plurality of access transistor pairs electrically connecting or disconnecting a corresponding bit line pair among the plurality of bit line pairs and a source line, in response to a voltage applied through a corresponding search line pair among a plurality of search line pairs according to data to be written or searched.

A four transistor ternary content-addressable memory cell includes a first series connection of a first non-hysteretic transistor (e.g., a thin-film transistor) and a first memory transistor (e.g., a thin-film transistor) including a first memory element configured to store a first binary bit; and a second series connection of a second non-hysteretic transistor and a second memory transistor including a second memory element configured to store a second binary bit. The first series connection and the second series connection are connected in parallel between a match line and a word line.

The invention relates to the field of integrated circuits, and discloses a reconfigurable 10T SRAM (Static Random Access Memory) unit circuit for energy-saving CAM (Content Addressable Memory) operation and memory calculation. The 10T SRAM circuit comprises reading ports which are decoupled in the horizontal direction and the vertical direction, supports cross-row and cross-column reading operation, and realizes matrix transposition, logic operation and parallel search functions. The invention provides a self-suppression discharge mechanism, the discharge swing of the matching line is obviously reduced, the sneak current problem caused by multi-row activation is effectively solved, and the anti-interference capability of the circuit is improved. Aiming at storage and calculation requirements, the invention provides three working modes, namely a conventional SRAM (Static Random Access Memory) mode, an in-memory logic operation mode and a content addressable memory mode. Especially in a content addressable memory mode, line-by-line search is adopted, additional data movement caused by vertical data storage is avoided, and compatibility among different working modes is enhanced.

The present invention provides a TCAM unit, belonging to the field of semiconductor integrated circuit design. The TCAM unit of the present invention includes: a CMOS field-effect transistor; a complementary CMOS field-effect transistor; a diode; a complementary diode; a memory cell, whose first end is electrically connected to the drain of the CMOS field-effect transistor and the cathode of the diode and can be in a high resistance state and a low resistance state; a complementary memory cell, whose first end is electrically connected to the drain of the complementary CMOS field-effect transistor and the cathode of the complementary diode; a bit / select line, which is electrically connected to the second end of the memory cell; a complementary bit / select line, which is electrically connected to the second end of the complementary memory cell; a match line, which is electrically connected to the diode and the anode of the complementary diode; a word line, which is electrically connected to the gates of the CMOS field-effect transistor and the complementary CMOS field-effect transistor; and a ground line, which is electrically connected to the source of the CMOS field-effect transistor and the complementary CMOS field-effect transistor. The TCAM unit of the present invention has low leakage current, low power consumption in operation and standby mode, fewer connection lines, and high storage density.

A recursive Analog Content Addressable Memory (ACAM) device includes a first comparison stage and a second comparison stage. The first comparison stage includes a first match line, a first base segment comparator connected between the first match line and a reference node, a first incremented segment comparator, and a first pass transistor. The first pass transistor and the first incremented segment comparator are connected in series between the first match line and the reference node. The second comparison stage includes a first inverter connected to the first pass transistor, a second match line connected to the first inverter, and a second base segment comparator connected between the second match line and the reference node. The device enables efficient comparison of multi-bit values.

A graph cache (200) comprising the following: a content-addressable memory array, CAM array, (202) comprising CAM rows (210) and matching lines corresponding to the CAM rows, wherein the CAM array is configured to receive an identifier of a target node of a graph structure (300), search for the identifier in the CAM rows and activate those of the matching lines that correspond to the CAM rows which store the identifier; a random access memory (RAM) array (208) comprising RAM lines (214) and word lines corresponding to the RAM lines; and a multiple match resolver (206) connected to the CAM array and the RAM array, wherein the multiple match resolver is configured to serially activate the word lines of the RAM array corresponding to the activated match lines of the CAM array.

The present disclosure discloses a content addressable memory apparatus. A self-test control circuit generates self-test driving signals and control an external test pattern generation circuit generates search data and an address signal. A content addressable memory circuit includes a memory circuit, an internal test pattern generation circuit and an internal comparison circuit. The memory circuit receives the search data and the address signal from the external test pattern generation circuit to perform search operation to generate actual match signals corresponding to the match lines. The internal test pattern generation circuit receives the self-test driving signals from the self-test control circuit to generate expected match signals. The internal comparison circuit receives and compares the actual match signals and the expected match signals to generate match comparison results such that the self-test control circuit receives the match comparison results and performs verification accordingly.

A photonic content addressable memory (CAM) (100) is provided. The photonic CAM (100) comprises: N search lines (101) and a plurality of match lines (102), each match line comprising N photonic CAM cells (110). Each search line (101) is configured to receive a search optical signal and each search optical signal comprises a different wavelength encoding one of N parallel search symbols. Each match line (102) is configured to receive: a wavelength division multiplexed content optical signal comprising a plurality of wavelengths, each wavelength encoding one of N parallel content symbols; and the N search optical signals from the N search lines (101). Each photonic CAM cell is configured to compare a respective search symbol with a respective content symbol and provide an output optical signal that is responsive to the comparison. Each match line (102) also comprises a content optical demultiplexer (120), configured to demultiplex the wavelength division multiplexed content optical signal and provide each CAM cell (110) with the respective content symbol.

The invention provides a gate controlled thyristor and a content addressable memory array. The grid control thyristor comprises a first transistor, a second transistor and a third transistor which are respectively provided with a control end, a first end and a second end. Control terminals of the first and third transistors are connected to a search line. The control end of the second transistor is connected to a fixed bias, and the second end is connected to the first end of the first transistor. The first end of the third transistor is connected to the matching line, and the second end is connected to the first end of the second transistor. A search bit is determined based on a search line voltage applied to the search line, and the search bit is compared with the data to determine whether the search bit is matched with the data. The architecture is suitable for a 3D NAND memory, and has the characteristics of high capacity and high performance.

A 3D search engine receives searches for application to word lines of a nonvolatile memory array. The engine uses two word lines per bit of information of the searches and two memory devices per bit of stored feature to search against, optionally enabling don't care and / or wildcard encoding. The engine uses respective bit lines of the nonvolatile memory array as respective matching lines for searching. Respective memory strings (e.g., NAND memory strings) of the nonvolatile memory array are usable to store respective data words, e.g., corresponding to features to search for. Respective pluralities of the memory strings are coupled in parallel to respective shared bit lines. Various encodings of features and searches enable exact, approximate, and range matching. The engine has applicability to comparing and sorting, in addition to searching in application areas such as artificial intelligence (AI) and big data.

The present invention discloses a ternary content addressable memory, which includes a search line and a match line connected to a TCAM cell; the match line is equipped with a charge sharing circuit; the charge sharing circuit includes: a metal line disposed on one side of the match line, and a charge sharing control unit for controlling the connection and disconnection between the metal line and the match line; the charge sharing control unit connects the metal line and the match line after the match line is precharged to a high level and before the search line is activated, so that charge sharing occurs between the match line and the metal line, and the voltage of the match line is reduced through charge sharing. The present invention can reduce the voltage of the match line through charge sharing before the search line is activated. After the voltage of the match line is reduced, the evaluation speed of the match line can be increased, the efficiency of the search operation can be improved, and thus the working frequency and performance of the ternary content addressable memory can be improved.

Systems and methods are provided for implementing a low power and area ternary content addressable memory (TCAM). An example of a TCAM comprises a match line, and a plurality of TCAM cells connected along the match line. Each TCAM cell stores a state of a threshold value. The TCAM cells are configured to pull down a signal over the match line in response to inequality between an input search and the threshold value. The plurality of TCAM cells comprises a number of TCAM cells that is less than the threshold value. The input values can be encoded according to a first encoding scheme and the threshold value can be encoded according to one of a second and a third encoding scheme based on an inequality check mapped to the plurality of TCAM cells.

The present disclosure provides a three-dimensional memory device and a ternary content-addressable memory cell thereof. The ternary content-addressable memory cell includes a first memory cell, a second memory cell, a first search switch, and a second search switch. The first memory cell is disposed in a first AND-type flash memory cell row. The second memory cell is disposed in a second AND-type flash memory cell row. The first search switch is coupled between a first bit line and a match line corresponding to the first AND-type flash memory cell row, and the first search switch is controlled by a first search signal to be turned on or off. The second search switch is coupled between a second bit line and a match line corresponding to the second AND-type flash memory cell row, and the second search switch is controlled by a second search signal to be turned on or off.

A method of storing datasets for multiple parties having a number of fields and line items wherein each line item contains some fields that relate to sensitive information of data subjects are obfuscated and also having non-obfuscated fields, wherein two parties are able to connect with and authorise the use of their data with the other and obfuscating the non-sensitive information of the data subjects of the first party, loading the second party data and matching the data sets to match line items from each dataset using the fields in order to create a third data set such that the first party can view the non-obfuscated data of the third data set, export said data, and analyse / display said data using a web tool. The parties may be able to include detailed criteria on what portion of data the first party can access, the criteria may include one or more defined fields and line items that are allowed to be accessed and whether the line items from all data subjects can be accessed or only those relating to similar data subjects.

In an implementation, a device includes: a first inverter; a first sense amplifier connected to the first inverter; a first match line connected to the first sense amplifier; a first analog content addressable memory cell including: a first pull-down transistor; a second pull-down transistor, the first pull-down transistor and the second pull-down transistor being connected in series between the first match line and a reference node; a first memory circuit configured to store a first upper bound of a first range and to activate the first pull-down transistor when an analog input value is less than the first upper bound of the first range; and a second memory circuit configured to store a first lower bound of the first range and to activate the second pull-down transistor when the analog input value is greater than the first lower bound of the first range.

A gate-controlled thyristor (GCT) and CAM memory are provided. The GCT includes first, second and third transistors, each having a control end, first and second ends. The control ends of the first and second transistors are connected to a search line. The control end of the second transistor is applied with a fixed bias voltage, and the second end of the second transistor is connected to the first end of the first transistor. The second end of the third transistor is connected to the first end of the second transistor, and the first end is connected to a match line. Based on the search line voltage, a search bit is determined to compare the data and the search bit to determine whether the search bit matches the data. This configuration may be applied to 3D NAND memory, having a high capacity and performance.

Examples of the presently disclosed technology provide CAM-based circuits specially constructed to implement Boolean satisfiability problems involving k-XOR-SAT clauses. With the strategic addition of auxiliary counting and logic circuits that evaluate match line voltage outputs of a CAM at k discrete times in order to determine whether a counted number of matches returned by a match line satisfies a pre-determine parity condition—where k represents a number of literals of a k-XOR-SAT clause of a Boolean satisfiability problem—a circuit of the present technology can leverage a common CAM (i.e., the same CAM) to implement the k-XOR-SAT clause and k-SAT clauses. Accordingly, this extremely versatile circuit can be used to implement k-XOR-SAT and k-SAT-k-XOR-SAT hybrid problems in less time, and with less hardware and power consumption than existing hardware accelerators.

The invention discloses an in-memory search matching line, method and system which can be applied to the technical field of analog circuits, and the matching line comprises a wire, a reading module, a power module, a node capacitor, a charging module, an MOS tube and a parasitic capacitor. Wherein one side of the wire is sequentially connected in series with the MOS tube and the reading module; at least one flash memory unit string is hung on the other side of the wire; the parasitic capacitor is mounted on the wire between the flash memory unit string and the MOS tube; and the charging module, the node capacitor and the power supply module are sequentially mounted on the wire between the MOS tube and the reading module. Therefore, the power supply module is additionally arranged on the matching line to charge the node capacitor, the current on the lead is quantified through the node voltage rising speed on the node capacitor read in the matching stage, the Hamming distance between different matching degrees is increased, and the accuracy of in-memory search is further improved.

A layout of a semiconductor memory device includes a substrate and a ternary content addressable memory. The ternary content addressable memory is disposed on the substrate and includes a plurality of ternary content addressable memory bit cells, and at least two of the ternary content addressable memory bit cells are mirror-symmetrical along a symmetry axis. Each ternary content addressable memory bit cell includes two memory cells electrically connected to two word lines, respectively, and a logic circuit electrically connected to the memory cells. The logic circuit includes two first read transistors and two second read transistors. Each second read transistor includes a gate and two source / drain regions. The source / drain regions of the second read transistor are electrically connected to two match lines and a first read transistor, respectively. The word lines are disposed parallel to the match lines.

A Content-Addressable Memory (CAM) cell has a latch of cross-coupled inverters and transmission gates to bit lines. Each transmission gate has n-channel and p-channel transistors in parallel that both turn on during writing. A node in the latch is applied to a gate of a data transistor in series with a select transistor to discharge a match line when select and latch data mismatch. Second data and select transistors receive inverse data from the latch for a CAM but receive data from a second cell latch for a Ternary Content-Addressable Memory (TCAM) cell with two latches per pair of select lines. When mismatches occur, even cells discharge the match line using n-channel data and select transistors while odd cells charge a complement match line using p-channel data and select transistors. The total number of p-channel and n-channel transistors in the cell can be equal when using complementary match lines.

The present application provides a content addressable memory (CAM) device, a content addressable memory cell and a data search comparison method thereof. The content addressable memory device includes a plurality of content addressable memory strings and an electrical characteristic detection circuit coupled to the content addressable memory strings. When performing a data search, a search data is compared to a stored data stored in the content addressable memory strings. The content addressable memory strings generate a plurality of memory string currents. The electrical characteristic detection circuit detects the memory string currents or detects a plurality of match line voltages of a plurality of match lines coupled to the content addressable memory strings to generate a plurality of search results. The stored data and the search data are a range stored data and a single bit search data, or the stored data and the search data are a single bit stored data and a range search data.

The invention provides an operation system, an operation method thereof and a search memory. A non-volatile 3D memory search architecture for receiving a plurality of searches for application to a plurality of select lines and a plurality of word lines of a non-volatile 3D memory. This architecture uses two word lines per cell of multiple searched information, and two memory devices per cell of a storage feature for reverse search. This architecture uses a plurality of corresponding bit lines of a non-volatile 3D memory array as a plurality of corresponding match lines for search. A plurality of corresponding memory strings (e.g., a plurality of NAND memory strings) of a non-volatile 3D memory array may be used to store a plurality of corresponding data values, e.g., corresponding to a plurality of elements to be searched. The plurality of corresponding memory strings are coupled in parallel to the plurality of corresponding common bit lines. In addition to searching in application areas such as artificial intelligence (AI) and big data, the architecture has application capabilities of comparison and sorting.

Differential content-addressable memory, dCAM, array (1600), comprising: a plurality of rows (1610) and columns (1615) of dCAM cells (1500, 1605), each row of the dCAM cells containing a high-match line (1511) and a low-match line (1510); a detection circuit (1620) electrically connected to the high-match line and the low-match line, the detection circuit comprising: a transimpedance amplifier, TIA, (1730) configured to detect a match line current on the low-match line, and an analog-to-digital converter, ADC, (1720) configured to detect an output voltage of the TIA and a match line voltage of the high-match line; and a digital-to-analog converter, DAC, (1625), which is electrically connected to each column of the dCAM cells.

The invention relates to the technical field of memories, and discloses a Boolean operation method based on a flash memory addressable memory. The two flash memory units M1 and M2 form a CAM unit for logical operation of n input variables, and threshold voltages Vth1 and Vth2 stored in the M1 and the M2 are used for representing the input variables X1, X2,..., Xn-1; the search voltages Vs1 and Vs2 applied by the grid electrodes of the M1 and the M2 are used for representing an input variable Xn, so that logic operation is realized; various logical operations are realized on one CAM unit according to a corresponding mapping rule through different mappings from an input variable logic value to a characteristic value of the CAM unit and mapping from a state of a matching line ML to an output result. According to the method, the reconfigurability and functionality of the CAM are improved.

The invention discloses an MRAM (Magnetic Random Access Memory)-based content addressable memory circuit and a pipeline search method. The circuit comprises a first node; a second node; the differential comparison circuit is connected with the first node and the second node, is used for connecting a search line and a search complementary line and storing complementary data, and is also used for performing differential discharge on the first node and the second node according to the voltage of the search line and the voltage of the search complementary line and generating differential voltage; the input end of the inverter latch circuit is connected with the output end of the differential comparison circuit, and the inverter latch circuit is used for amplifying the differential voltage and then outputting the amplified differential voltage; the transmission matching circuit is used for being connected with a matching line, the input end of the transmission matching circuit is connected with the output end of the inverter latch circuit, and the transmission matching circuit is used for outputting a mismatch signal when it is detected that the complementary data is not matched with the search data according to the differential voltage. The content addressable memory circuit based on the MRAM solves the problem of high search error rate of the content addressable memory circuit based on the MRAM.

Circuit (600) which has the following features: an analog content-addressable memory aCAM array (610) comprising a row of aCAM cells (610 (A) - (Y)) electrically connected along a matching line, each aCAM cell from the row of aCAM cells being programmed to store an analog voltage threshold representing an entry of a word programmed representatively across the row of aCAM cells, wherein: the row of aCAM cells contains a redundancy aCAM cell (610(A)(n-1) - 610(Y)(n-1)) which is programmed to store a calculated redundancy analog voltage threshold, and The redundancy analog voltage threshold is calculated such that each row of a binary array representing the word in binary digits contains a sequence of ones and zeros specified by a binary parity check matrix.