42 results about "Auxiliary memory" patented technology

The invention relates to the technical field of memory management, in particular to a dynamic memory management method based on a bidirectional queue, electronic equipment and a medium. A first pointer and a second pointer are set to orderly control bidirectional insertion or deletion of queue elements, a queue main memory is started firstly, and when the storage requirement of the queue elements is not met, areas stored with the queue elements are reordered and then a queue auxiliary memory is started; when the queue main memory and the queue auxiliary memory are not enough to meet the storage requirements of the queue elements, updating the sizes of the queue main memory and the queue auxiliary memory to expand the queue memory, reordering discrete regions, in which the queue elements are stored, in the original queue memory, generating a continuous region, and writing the continuous region into a new queue main memory; the frequency of queue element copying and memory releasing in the dynamic change process of the bidirectional queue is reduced, and the memory utilization rate of the bidirectional queue is increased.

A method of operating a ring communication network, including providing a management device having a first processor, a first memory, a first communicator, and a computer program stored in the first memory and capable of running on the first processor; providing a network device having a network processor, a network memory, and a network communicator; disposing an auxiliary memory in the network device; organizing the management device and the network device to form the ring communication network; determining a location of the auxiliary memory in the ring communication network; and assigning a transmission direction such that a transmission distance between the management device and the auxiliary memory is minimized.

Embodiments of the present disclosure relate to a method of protecting a system such as a microcontroller and a corresponding system. A system comprising a processing unit, a memory configured to store at least one first set of instructions and one second set of instructions for execution by the processing unit, the processing unit being configured to sequentially fetch instructions of the first set and instructions of the second set from the memory for their execution. The system further comprises a controller comprising a first auxiliary memory configured to store a protection criterion, a comparator configured to compare a memory address of each fetched instruction with the protection criterion, and a control circuit configured to trigger a protection mechanism in response to the memory address satisfying the protection criterion, the protection mechanism comprising at least one prohibition of the processing unit to execute again at least a portion of the instructions of the first set during the execution of the instructions of the second set.

This increases the likelihood that storage at a tier relatively close to the computing resources is holding the data at the time it is used to execute a job. [Solution] Before starting the execution of job 102, the job analysis unit 300 analyzes the workflow of job 102 to identify the data that the computing resource 210 will read when executing job 102. The pre-read management unit 1200 controls the pre-reading from secondary storage 240 to primary storage 230 for data identified by the job analysis unit 300 for job 102 where the actual data 920 does not exist in primary storage 230 but does exist in secondary storage 240.

This invention provides a management system for controlling the audio information output from an audio generator when multiple target objects are used separately at multiple locations. [Solution] A management system 10 comprising a first target object 13 and a second target object 14, a user terminal 12, a speaker and audio output device 43 for outputting sound, and a management computer 11 that is communicatively connected to the user terminal 12 and performs processing, wherein the management computer 11 is configured to perform the following processes: storing the first target object 13 and the second target object 14, the first usage location A1 and the second usage location A2 in an auxiliary memory 49; determining which of the first usage location A1 and the second usage location A2 is being used; determining the usage location of the used target object; and controlling the audio information output from the speaker and audio output device 43 based on the determination result of the target object and usage location being used.

To provide an information managing device that facilitates the development of a relationship between a user and the character of the user.SOLUTION: An information managing device 10 includes a processor 19 that executes processes, and is connected to a user terminal 12 via a network 22. This information managing device 10 is configured to cause the processor 19 to execute: a first process of storing, in an auxiliary memory 21, the personality and activity of a character which are generated in accordance with activity information on the character that shows up in a first action application which is run upon operation given to the user terminal 12 from a user 13; and a second process of taking over, as the personality and activity of the character that shows up in a second action application which is run after the first process is executed, and which has different functions from those of the first action application, the personality and activity of the character stored in the auxiliary memory 21 in the first process.SELECTED DRAWING: Figure 1A

When a memory controller reads target data in a memory, the memory controller reads the target data and first check code of the target data from the memory. The memory controller checks the target data using the first check code. If the check fails, it indicates that error data exists in the target data. The memory controller then performs error correction on the target data using the first check code. After failing to perform error correction on the target data using the first check code in the memory, the memory controller performs error correction on the target data using the second check code in the secondary storage.

System, comprehensive: an embedded modem with an internal memory whose size is too small to store the received data, which communicates with controllers via a vehicle bus, each controller including a working memory and an auxiliary memory, wherein the modem is programmed to to query the controllers via the bus in order to identify controllers that have available memory in the auxiliary memory, If the available memory is located in the auxiliary memory of a priority controller, the received data is sent to the priority controller via the bus, and Otherwise, if the available memory in the auxiliary memory of a subordinate controller is available, the received data should be sent to the subordinate controller via the bus. to query the controllers via the bus to confirm that the controller storing the received data is available for data storage, and to confirm that the controllers are not operating in a driving mode, and in response to the confirmation, to request the stored data via the bus and to transmit the stored data to a telematics server via a wide area network; the embedded modem is further programmed, if no available memory is found, to reduce the data to a size corresponding to the internal memory of the embedded modem.

A printing system includes a printing device that receives print jobs. The printing device includes a controller having a raster image processing (RIP) system that includes a RIP manager and at least one renderer. The RIP manager detects that the print engine of the printing device is idle. The RIP system renders various jobs using renderers during the idle time and stores the rendered pages in a storage for the RIP system. Once a storage threshold is reached, the RIP system allocates the renderers to render only complex pages within the print job and stores the rendered complex pages to a secondary storage. If a secondary job is active, then available renderers are allocated to rendering pages of the secondary job. Once printing resumes at the print engine, the RIP manager reallocates the rendering resources back to normal.

The disclosure provides a system and method for reducing inference latency of an artificial intelligence (AI) system. During operation, the system can obtain an AI model and compile the AI model to generate at least one Directed Acyclic Graph (DAG), which comprises determining an offset address associated with a piece of intermediate data to be transferred from a primary memory shared among multiple AI accelerators to a secondary memory. The AI accelerators, the primary memory, and the secondary memory are located on the same system on a chip (SoC). The system can then schedule computing tasks, which comprises determining a base address associated with the DAG in the secondary memory, and perform inference based on the DAG, which comprises transferring the piece of intermediate data from the primary memory to the secondary memory based on the offset address and the base address.

This disclosure provides methods, components, devices and systems for device coordinated dynamic configuration management for high reliability communication. Some aspects more specifically relate to techniques to coordinate the transfer of data between primary and secondary memories of a device such as an access point (AP) for communication with a wireless station (STA). The AP may be capable of storing data in a directly accessible primary memory, and in a secondary memory, which may be inaccessible until the data in the secondary memory is transferred to the primary memory. In some aspects, the AP may dynamically transfer data based on changes in one or more operating modes of a STA. The AP may transfer data from the secondary memory to the primary memory in accordance with the STA entering an active mode, or from the primary memory to the secondary memory in accordance with the STA entering a power saving mode.

Computing device (100) comprising a main memory (104) configured to store a sparse matrix in a dense vector format (106, 108, 110) and to store a second vector (112) or a second matrix, a computing unit (102) configured to multiply the sparse matrix by the second vector or by the second matrix, and a streamer (114) comprising: an indexed loading block (116) comprising a secondary memory (118) and a FIFO memory (120) for requests to send values stored in the secondary memory to the computing unit; an indexed loading engine (122) configured to sequentially generate and store requests in the FIFO request memory according to an order in which the values are intended to be sent to the computing unit; the request storage order being calculated and stored in the form of firmware (124).

The invention discloses a semiconductor device, a preparation method thereof and electronic equipment. The semiconductor device comprises a substrate, a plurality of storage units and at least one auxiliary storage layer, the storage layers of two adjacent storage units are arranged at intervals in the direction perpendicular to the substrate; each auxiliary storage layer is located between every two adjacent storage units and located on the periphery of the corresponding channel structure, and the auxiliary storage layers and the storage layers are arranged in a staggered mode in the direction perpendicular to the substrate. The invention relates to but not limited to the technical field of semiconductor devices, in particular to a semiconductor device, a preparation method thereof and electronic equipment, and can solve the problem of coupling between adjacent layers of storage units.

Computer computing device with memory optimized for matrix calculation. Computer computing device (100) comprising a main memory (104) configured to store a sparse matrix in a dense vector format (106, 108, 110) and to store a second vector (112) or a second matrix, a computing unit (102) configured to multiply the sparse matrix by the second vector or by the second matrix, and a loading circuit (114) comprising: - an indexed loading block (116) having a secondary memory (118) and a FIFO query memory (120) for sending values ​​stored in the secondary memory to the computing unit; - an indexed loading engine (122) configured to sequentially generate and store the queries in the FIFO query memory in an order in which the values ​​are intended to be sent to the computing unit;the order in which queries are stored is calculated and stored in the form of a microprogram (124). Figure for the abbreviation: Fig. 1;

A medical visualisation system including a medical visualisation device for performing a medical visualisation procedure and a monitor device. The medical visualisation system includes an auxiliary component including an auxiliary memory and an auxiliary communication interface. The auxiliary memory storing initial data and the auxiliary component being configured to transmit the initial data to the monitor device using the auxiliary communication interface, and wherein a monitor processing unit of the monitor device is configured to receive the initial data and adjust one or more parameters of the medical visualisation system based on the initial data.

The invention relates to the technical field of data processing, and discloses an auxiliary memory material generation method and device, computer equipment and a medium, and the method comprises the steps: obtaining target information inputted by a user; the target information comprises concept information used for describing to-be-memorized information; generating a plurality of target visual images corresponding to the concept information; the target visual image comprises content for assisting visual associative memory; based on the target interaction mode, combining the target visual images with the interaction relationship from the multiple target visual images to obtain a visual image combination; generating a scene description corresponding to each visual image combination; an auxiliary memory material is generated based on the scene description. By adopting the method, the generated auxiliary memory material can help the user to master the concept information more efficiently. Moreover, the manufacturing cost can be greatly reduced, and manpower, material resources and time are saved, so that the high-quality auxiliary memory material is quickly generated, and the large-scale application requirement is met.

A method of operating a ring communication network, including providing a management device having a first processor, a first memory, a first communicator, and a computer program stored in the first memory and capable of running on the first processor; providing a network device having a network processor, a network memory, and a network communicator; disposing an auxiliary memory in the network device; organizing the management device and the network device to form the ring communication network; determining a location of the auxiliary memory in the ring communication network; and assigning a transmission direction such that a transmission distance

The invention relates to an auxiliary memory method and system based on machine learning, and belongs to the field of machine learning, and the method comprises the steps: setting the length of a user input bit according to the length of a target memory word letter sequence; obtaining a memory feedback text input by the user; the Hamming distance between the target memory word letter sequence and the memory feedback letter sequence is calculated, and if the Hamming distance is larger than 0, the corresponding position of the target memory word letter sequence reckoned in the Hamming distance is marked as a forgetting position; and when the user interaction module displays the target memory word, the letters of the forgotten positions of the target memory word are displayed differently. According to the method and the device, the word memory effect can be distinguished more finely, the letters of the forgotten positions in the word are displayed differently in combination with the forgotten positions, the user is helped to focus on the part which is easier to forget, and the auxiliary memory effect is improved.

A job analysis unit analyzes a workflow of a job before starting execution of the job to specify data to be read when a calculation resource executes the job. A prefetch management unit performs control to start prefetch from a secondary storage to the primary storage for data of which no data entities exist in the primary storage and data entities exist in the secondary storage, among the data specified by the job analysis unit for the job.

Device and computer-implemented method for determining a memory plan (200) for performing operations, in particular of an artificial neural network, wherein a list of memory areas required for performing the operations is created, wherein, depending on the list for the operations, it is determined which memory areas must be present in a first memory (202) for performing the respective operation and which memory areas may be present in a second memory (204) during the execution of the respective operation, wherein the memory plan (200) is determined depending on whether a memory area must be present in main memory for execution or may be present in secondary memory.

[Problem] To provide a management system capable of controlling voice information to be outputted from a voice generation device when a plurality of target objects are separately used in a plurality of use places. [Solution] A management system 10 comprises: a first target object 13 and a second target object 14; a user terminal 12; speakers 23, 32 and a voice output device 43 that output voice; and a management computer 11 that is communicably connected to the user terminal 12 and executes processing. The management system 10 is configured such that the management computer 11 executes processing for storing the first target object 13, the second target object 14, a first use place A1, and a second use place A2 in an auxiliary memory 49, processing for determining which of target objects in the first use place A1 and the second use place A2 is used, processing for determining the use place of the used target object, and processing for controlling voice information including voice to be outputted from the speakers 23, 32 and a the voice output device 43 on the basis of the determination result of the used target object and the determination result of the use place.

An apparatus and computer-implemented method for determining a memory plan for executing operations, in particular of an artificial neural network. A list of memory areas required for executing the operations is created, wherein, depending on the list, it is determined for the operations which memory areas must be present in a first memory for executing the particular operation and which memory areas may be present in a second memory during the execution of the particular operation, wherein the memory plan is determined depending on whether a memory area must be present in the main memory for execution or may be present in the secondary memory.

The application relates to the technical field of memory management, in particular to a dynamic memory management method based on a one-way queue, electronic equipment and a medium. The application stores queue elements in the mode of setting a queue main memory and a queue auxiliary memory, first activates the queue main memory, then activates the queue auxiliary memory, and when the queue main memory and the queue auxiliary memory cannot meet the storage requirements of the queue elements, the size of the queue main memory and the queue auxiliary memory is updated to expand the capacity, the frequency of copying and releasing the memory of the queue elements in the dynamic change process of the one-way queue is reduced, and the queue memory utilization rate of the one-way queue is improved.

The invention discloses an anomaly detection method and electronic equipment, and the method comprises the steps: determining a plurality of features of a to-be-detected image, a group of first similarities between the features of the to-be-detected image and a plurality of features in a first-stage memory module, and at least one group of second similarities between the features of the to-be-detected image and a plurality of features in at least one-stage auxiliary memory module; the primary memory module is used for storing normal characteristics; the auxiliary memory module is used for storing fuzzy features between normal features and abnormal features; fusing the group of first similarities and the at least one group of second similarities to obtain a plurality of fusion similarities; and determining an abnormal region in the to-be-detected image according to the plurality of fusion similarities.

The invention relates to a read-only memory array structure, a read-only memory, a reading method of the read-only memory and electronic equipment. The read-only memory array structure comprises N * M memory units, wherein N and M are integers greater than or equal to 2; wherein the grid electrode of the NMOS transistor of at least one auxiliary storage unit in the N * M storage units is connected to the gating word line of the target storage unit in the N * M storage units, and the drain electrode of the NMOS transistor of at least one auxiliary storage unit is connected to the target word line of the target storage unit; wherein the auxiliary storage unit is a storage unit for storing data 1, and the target storage unit is a storage unit for storing data 0. According to the read-only memory array structure, design optimization is carried out on the read-only memory array structure, when the target memory unit is read, the NMOS transistor of the at least one auxiliary memory unit is started, the NMOS transistor of the at least one auxiliary memory unit and the NMOS transistor of the target memory unit form a parallel pull-down path, discharging is carried out on the target bit line together, and therefore the pull-down capacity is enhanced.

A gaming machine is described herein. The gaming machine includes a control unit including a control board, a primary memory device coupled to the control board, a secondary memory device coupled to the control board, and a processor. The processor is programmed to execute an algorithm including the steps of initiating an instance of a primary game by spinning and stopping reels to display an outcome of the primary game. The processor then detects a trigger condition appearing in the outcome of the primary game and responsively displays an instance of a bonus game, generates primary game recall data including instructions for displaying the outcome of the primary game and stores the primary game recall data in the primary memory device, and generates bonus game recall data including instructions for displaying the instance of the bonus game and stores the bonus game recall data in the secondary memory device.

The invention relates to the technical field of memory management, in particular to a dynamic memory management method based on a one-way queue, electronic equipment and a medium. Queue elements are stored by setting a queue main memory and a queue auxiliary memory, the queue main memory is started firstly, and then the queue auxiliary memory is started; and when the queue main memory and the queue auxiliary memory are not enough to meet the queue element storage requirement, the sizes of the queue main memory and the queue auxiliary memory are updated to expand the capacity, so that the frequency of copying the queue elements and releasing the memories in the dynamic change process of the one-way queue is reduced, and the queue memory utilization rate of the one-way queue is improved.

There is provided a data processing apparatus and method. The data processing apparatus comprises a plurality of processing elements connected via a network arranged on a single chip to form a spatial architecture. Each processing element comprising processing circuitry to perform processing operations and memory control circuitry to perform data transfer operations and to issue data transfer requests for requested data to the network. The memory control circuitry is configured to monitor the network to retrieve the requested data from the network. Each processing element is further provided with local storage circuitry comprising a plurality of local storage sectors to store data associated with the processing operations, and auxiliary memory control circuitry to monitor the network to detect stalled data (S60). The auxiliary memory control circuitry is configured to transfer the stalled data from the network to an auxiliary storage buffer (S66) dynamically selected from amongst the plurality of local storage sectors (S64).

Embodiments of the present disclosure disclose a storage and retrieval method for memory metadata for large model inference assistance. A specific implementation of the method includes: obtaining multi-modal perception information and historical session interaction context information corresponding to a session identifier; generating short-term auxiliary memory metadata, and storing the short-term auxiliary memory metadata into a short-term memory meta space constructed by a static random access memory; dynamically optimizing and transferring each short-term auxiliary memory metadata in the short-term memory meta space; generating knowledge inference auxiliary memory metadata; performing progressive forgetting dynamic compression processing on at least one multiplexing compression memory metadata in a long-term memory meta space; in response to determining that an inference task is started, retrieving based on session interaction information to obtain a multi-level memory metadata set and a candidate knowledge memory metadata set for a large model to generate and feed back inference reply information to a user terminal. The implementation improves the real-time response speed of the inference task.

A method for selectively copying data of a primary memory, wherein the data is stored on one or more blocks of one or more logical units of the primary memory. The method includes receiving, by a secondary memory, a request to selectively copy the data of the primary memory. The method also includes executing a command to receive a configurable set of parameters of the logical unit. The method also includes selectively copying, from the primary memory to the secondary memory, the one or more blocks storing the data in accordance with the received configurable set of parameters.