Core particles, multi-core particle systems, and methods for core particles
Patent Information
- Application Number
- CN202580018462.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-03-01
- Filing Date
- 2025-03-03
- Publication Date
- 2026-09-29
Smart Images

Figure CN122847701A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to a core for processing data associated with at least one data source, such as a video data source.
[0002] This disclosure also relates to a multi-core system.
[0003] This disclosure also relates to a method for use with a chip. Summary of the Invention
[0004] Some embodiments relate to a core for processing data associated with at least one data source, wherein the core is configured to exchange the data with at least one other device, such as at least one other core, according to a streaming media protocol. In some embodiments, this enables the efficient exchange of information, such as information associated with image data or video data.
[0005] In some embodiments, a chip is an integrated circuit having a predefined function or sub-function. In some embodiments, a chip may be combined with at least one other chip, for example, combined on a common substrate, to form a multi-chip system.
[0006] In some embodiments, the data includes at least one of the following: a) image data, or b) video data. In other embodiments, information of a different type than the types mentioned for the foregoing examples, such as data, may also be exchanged by the core according to a streaming media protocol.
[0007] In some embodiments, a streaming media protocol is a protocol, such as a standard, that can be used to transmit data, such as multimedia data like video and / or audio, and that the protocol enables, for example, the continuous transmission and / or reception and / or parsing of content, such as “playing” or processing, such as further processing, without the need for the entire dataset, such as the entire file, to have already been transmitted.
[0008] In some embodiments, the exchange performed according to the streaming media protocol has at least one of the following: a) exchanging the data by using, for example, an existing, standardized streaming media protocol, or b) exchanging the data based on the streaming media protocol, wherein, for example, according to some embodiments, at least one aspect of an existing streaming media protocol is extended, for example, regarding the application of the protocol on a chip.
[0009] In some embodiments, the core is configured to process the data, for example, at least temporarily during or after data exchange.
[0010] In some embodiments, the processing has at least one of the following: a) processing a portion of the data, for example before fully receiving the data, for example according to a streaming media protocol, and / or during the transmission of the data, for example according to a streaming media protocol, wherein, for example in the case of image data, the portion of the data represents and / or includes at least one of the following: a1) a row, or a2) an image element, such as a pixel, or a3) a tile.
[0011] In other words, in some embodiments, it may be specified that data exchange according to a streaming media protocol includes exchanging, for example, sending and / or receiving multiple data units or multiple parts, such as multiple datasets, or, in the case of image data or video frames, exchanging, for example, rows, pixels or tiles (e.g., “tiles”).
[0012] In some embodiments, portions of a frame, for example, can be exchanged according to a streaming media protocol, i.e., can be received by a kernel, wherein, for example, after receiving a portion of a frame, i.e., while the kernel is receiving one or more other portions of the same frame, information associated with the corresponding portion of that frame can already be processed by the kernel. This provides the advantage that the kernel receiving the frame does not need to wait to fully receive the frame before it can begin processing the information of that frame. Thus, in some embodiments, efficient processing can be achieved, for example, processing that overlaps at least partially in time with, for example, data reception, thereby reducing latency in some embodiments.
[0013] In other embodiments, streaming media protocols can be used for the chip to receive multiple portions of an image, such as a digital image, for example, sequentially. For example, a data source can transmit a line or a tile of an image sequentially to the chip using a streaming media protocol, and, for example, after receiving the first line or the first tile of the image, the chip can process the information represented by the first line or the first tile, for example, without having received the second line or the second tile, and so on.
[0014] In some embodiments, the principles described above based on multiple embodiments may also be applied accordingly to other portions of image data or frames or video data, for example, for receiving and / or sending data according to a streaming media protocol.
[0015] In some embodiments, the chip is configured to perform at least one of the following: a) receiving the data from the at least one other device, for example from the at least one other chip, wherein, for example, the at least one other device is disposed on the same substrate as the chip; or b) receiving the data from at least one data source, wherein, for example, the at least one data source is not disposed on the same substrate as the chip; or c) transmitting the data to the at least one other device, for example, to the at least one other chip.
[0016] In some embodiments, the core has a memory, such as a buffer memory, such as a working memory, for example for at least temporarily storing at least a portion of the data.
[0017] In some embodiments, the chip has a communication system, such as an on-chip network or a cross switch, for exchanging data within the chip, such as exchanging data with a processor or computing core, and / or exchanging data with at least one external device, such as an external storage device.
[0018] In some embodiments, the chip has at least one processing unit configured to process image data and / or video data, for example, dedicated to processing image data and / or video data, such as having at least one image signal processor.
[0019] Some embodiments relate to multi-core systems having a plurality of cores, wherein at least one of the plurality of cores is constructed in accordance with the present disclosure.
[0020] In some embodiments, the multi-core system includes at least one of the following: a) a management core, for example for managing at least one aspect of the multi-core system; or b) a processing core having and / or characterizing at least one processing means, such as a computing device, such as a processor device, such as a central processing unit device, such as a CPU; or c) a graphics processing unit core, such as a GPU core; or d) an AI core for performing aspects of artificial intelligence, such as performing aspects of machine learning methods, wherein, for example, the AI core has a local storage means, for example for at least temporarily storing data for performing aspects of artificial intelligence, such as parameters, such as weights or hyperparameters, for at least one artificial neural network.
[0021] In some embodiments, a) the management chip and the AI chip are configured to exchange data with each other according to a streaming protocol, for example via a first data connection configured to perform a streaming protocol or a streaming protocol-based method; or b) the management chip and the graphics processor chip are configured to exchange data with each other according to a streaming protocol, for example via a second data connection configured to perform a streaming protocol or a streaming protocol-based method; or c) the management chip and the processing chip are configured to exchange data with each other according to a streaming protocol, for example via a third data connection configured to perform a streaming protocol or a streaming protocol-based method; or d) at least two of the chips are configured to exchange data with each other, for example directly with each other, for example without using the management chip, according to a streaming protocol, for example via a fourth data connection configured to perform a streaming protocol or a streaming protocol-based method.
[0022] Some embodiments relate to a product, such as a control device, for example a control device for a vehicle, such as a motor vehicle, comprising at least one of the following: a) a chip according to the present disclosure, or b) a chip system according to the present disclosure, or c) a device according to the present disclosure.
[0023] Some embodiments relate to a method for a kernel for processing data associated with at least one data source, the method comprising: exchanging data with at least one other device, such as at least one other kernel, according to a streaming media protocol.
[0024] Some embodiments relate to apparatus for performing methods according to the present disclosure.
[0025] Some embodiments relate to a computer-readable storage medium containing instructions that, when executed by a computer, cause the computer to perform a method according to the present disclosure.
[0026] Some embodiments relate to a computer program that includes instructions that, when executed by a computer, cause the computer to perform the method according to this disclosure.
[0027] Some embodiments relate to data carrier signals, the transmission of which and / or characterization are performed according to computer programs of this disclosure.
[0028] Some embodiments relate to using a chip and / or a multi-chip system and / or a product and / or a method and / or an apparatus and / or a computer-readable storage medium and / or a computer program and / or a data carrier signal according to the present disclosure for at least one of the following: a) exchanging data with at least one chip according to a streaming media protocol; or b) using aspects of a streaming media protocol for data exchange within a multi-chip system; c) using aspects of a streaming media protocol for data exchange between at least one component of a multi-chip system and at least one component or device located external to the multi-chip system; or d) reducing latency, such as latency in processing and / or playing, for example, image data and / or video data; or e) reducing the storage bandwidth requirements for at least one interface with respect to a chip and / or for at least one storage device for a chip and / or for a multi-chip system.
[0029] Other features, applications, and advantages of the invention can be derived from the following description of the embodiments of the invention shown in the accompanying drawings. Herein, all described or illustrated features, either alone or in any combination, constitute the subject matter of the invention, regardless of their general or figurative relationship in the claims, or their representation or illustration in the specification or drawings. Attached Figure Description
[0030] Figure 1 A simplified block diagram is shown schematically. Figure 2 A simplified flowchart is shown schematically. Figure 3 A simplified flowchart is shown schematically. Figure 4 A simplified block diagram is shown schematically. Figure 5 A simplified block diagram is shown schematically. Figure 6 A simplified block diagram is shown schematically. Figure 7 A simplified block diagram is shown schematically. Figure 8 A simplified block diagram is shown schematically. Figure 9 It schematically illustrates various aspects of the application. Detailed Implementation
[0031] Some embodiments, see Figure 1 , Figure 2The invention relates to a core 100 for processing data DAT-DQ associated with at least one data source DQ, DQ', DQ'', DQ''', wherein the core 100 is configured to exchange the data DAT-DQ with at least one other device 100', such as at least one other core 100', according to a streaming media protocol SP. In some embodiments, this enables the efficient exchange of information associated with, for example, image data or video data.
[0032] In some embodiments, a chip is an integrated circuit having a predefined function or sub-function. In some embodiments, a chip, for example according to... Figure 1 The core 100 can be combined with at least one other core 100', for example, on a common substrate (not shown), to form a multi-core system (see below, for example). Figure 4 , Figure 6 ).
[0033] In some embodiments, see Figure 1 The data DAT-DQ has at least one of the following: a) image data DAT-BILD, or b) video data DAT-VID. In other embodiments, information, such as data, of a different type than those mentioned in the foregoing embodiments, may also be exchanged by the chip 100 according to the streaming media protocol SP.
[0034] In some embodiments, see Figure 1 At least one data source DQ may have a data connection with core 100, but may not be integrated into core 100, for example. See other embodiments. Figure 1 A similar situation applies to data source DQ''', for example, regarding other cores 100'.
[0035] In some embodiments, see Figure 1 At least one data source DQ' can have a data connection with the core 100, for example, it can be integrated into the core 100. In other embodiments, see... Figure 1 A similar situation applies to data source DQ'', for example, regarding other cores 100'.
[0036] In some embodiments, the data connection DV between cores 100, 100', such as a bus system, can be used for data exchange 200 according to the streaming media protocol SP. Figure 2 In some embodiments, the data connection DV may be, for example, a UCIe (Universal Chiplet Interconnect Express) or a CXL (Compute Express Link) type.
[0037] In some embodiments, data exchange according to the streaming media protocol SP may also be performed with at least one other device, such as a data sink DS, wherein the other device DS is not configured as a core, for example.
[0038] In some embodiments, see Figure 2 The exchange 200 performed according to the streaming media protocol SP has at least one of the following: a) exchanging the data of 200a by using, for example, an existing, such as a standardized streaming media protocol, or b) exchanging the data of 200b based on the streaming media protocol SP, wherein, for example, according to some embodiments, at least one aspect of the existing streaming media protocol is extended, for example, regarding the application of the protocol on the chip 100.
[0039] In some embodiments, see Figure 1 , Figure 2 The core 100 is configured to process 202 data DAT-DQ, for example, at least temporarily during or after data exchange 200, wherein, for example, processed data DAT-DQ' is obtained.
[0040] In some embodiments, see Figure 2 The process 202 has at least one of the following: a) processing 202a a portion of the data DAT-DQ DAT-DQ_TEIL, for example before the data is fully received, for example according to the streaming media protocol SP, and / or during the transmission of the data, for example according to the streaming media protocol SP, wherein, for example in the case of image data DAT-BILD, the portion of the data DAT-DQ-TEIL represents and / or has at least one of the following: a1) a row, or a2) an image element, such as a pixel, or a3) a tile.
[0041] In other words, in some embodiments, data exchange according to the streaming media protocol SP may be specified to include exchange 200, such as sending and / or receiving multiple data units or multiple parts, such as multiple datasets, and in the case of image data or video frames, such as exchanging rows, pixels or tiles (e.g., “tiles”).
[0042] In some embodiments, therefore, portions of a frame may be exchanged, for example, according to the streaming media protocol SP, i.e., by chip 100 ( Figure 1The information associated with that portion of the frame can be processed by the core 100 simultaneously, for example, after receiving a portion of the frame, i.e., while the core 100 is receiving one or more other portions of the same frame. This provides the advantage that the core 100 receiving the frame can begin processing the information of the frame without waiting for the frame to be fully received. Thus, in some embodiments, efficient processing can be achieved, for example, at least in part in time with, for example, data reception (see, for example, see...). Figure 2 The boxes in the frame 200 overlap, thereby reducing latency in some embodiments, for example.
[0043] In other embodiments, see Figure 1 The streaming media protocol SP can be used for: the chip 100 to receive multiple parts of an image, such as a digital image, for example, to receive them sequentially.
[0044] For example, data sources DQ'' and 100' can transmit a line or a tile of an image to core 100 sequentially using the streaming media protocol SP. And, for example, after receiving the first line or the first tile of the image, core 100 can process the information represented by the first line or the first tile without having received the second line or the second tile, and so on.
[0045] In some embodiments, the principles described above based on multiple embodiments may also be applied accordingly to other portions of image data or frames or video data, for example, for receiving and / or sending data according to a streaming media protocol.
[0046] In some embodiments, see Figure 3 Core 100 ( Figure 1 The device is configured to perform at least one of the following: a) receiving the data DAT-DQ from the at least one other device, for example from the at least one other chip 100', wherein, for example, the at least one other device 100' is disposed on the same substrate as the chip 100; or b) receiving the data DAT-DQ from at least one data source, wherein, for example, the at least one data source is not disposed on the same substrate as the chip 100; or c) transmitting the data DAT-DQ to the at least one other device 100', DS, for example, to the at least one other chip 100'.
[0047] Figure 4 A simplified block diagram is schematically shown, illustrating the transmission of data from a first core 100a to a second core 100b according to some embodiments. For example, these two cores 100a and 100b constitute a multi-core system 1000. Figure 4The common substrate for these two cores is not shown in the figure.
[0048] In some embodiments, dataset D resides in a first core 100a and is to be transmitted to a second core 100b. In some embodiments, portions D-T1, D-T2, D-T3, ... of dataset D are sequentially transmitted from the first core 100a to the second core 100b via data connection DV' using a streaming media protocol SP. This transmission may be continuous or repeated until the entire dataset D has been received in the second core 100b. Because the transmission is performed using the streaming media protocol SP, the second core 100b may process the first portion D-T1 immediately after receiving it, for example, while the core 100b is waiting to transmit other portions D-T2, D-T3, or while the core 100b is receiving other portions D-T2, ...
[0049] In some embodiments, see Figure 4 The data connection DV' can be, for example, UCIe (Universal Chiplet Interconnect Express) or CXL (Compute Express Link) type. Optionally, in some embodiments, at least one additional control line SL can be provided between chips 100a and 100b, which can be used for signal transmission, for example, within the framework of using the streaming media protocol SP.
[0050] In some embodiments, see Figure 4 The dataset D represents an image, while the parts D-T1, DT-2, D-T3, ... represent, for example, individual rows, pixels, or groups of pixels in the image, such as tiles.
[0051] In some embodiments, see Figure 5 The chip 100b has a memory 110, such as a buffer memory, such as a working memory (e.g., RAM), for example for at least temporarily storing at least a portion of data, such as via a data connection DV' according to a streaming media protocol, such as from... Figure 4 The data received by the 100a chip.
[0052] In some embodiments, see Figure 5 The size of the buffer memory 110 is designed to accommodate at least a portion of the data D-T1, i.e., at least temporarily storing at least one row or one tile of, for example, a received image.
[0053] In some embodiments, see Figure 5The chip 100b includes a communication system 120, such as an on-chip network or a crossbar switch, for exchanging data within the chip 100b, such as exchanging data with a processor or computing core 130, and / or exchanging data with at least one external device 140, such as an external storage device.
[0054] In some embodiments, the data path of a portion D-T1 of dataset D, such as an image, that can be received and processed by core 100b may have, for example, the following stations: 1. Received and cached in memory 110 via streaming media protocol SP; 2. Transmitted to processor 130 via core internal communication system 120; 3. Transmitted from processor 130 via core internal communication system 120 to external memory 140, for example as a portion of dataset D processed by processor 130 (not shown).
[0055] In some embodiments, see Figure 5 The chip 100b has at least one processing unit 130 configured for processing image data and / or video data, for example, dedicated to processing image data and / or video data, and for example, has at least one image signal processor, such as an image signal processor (ISP). See also, in some embodiments. Figure 5 The processing unit 130 may, for example, be characterized by or have one or more ISPs.
[0056] See Figure 4 , Figure 6 Some embodiments relate to multi-core systems 1000, 1000a, the multi-core system having a plurality of cores, wherein at least one of the plurality of cores is constructed in accordance with the present disclosure.
[0057] For example, according to Figure 4 In the multi-core system 1000, both cores 100a and 100b can be constructed according to this disclosure.
[0058] In other embodiments, for example, it may be only one of the plurality of cores 100a, 100b of the multi-core system 1000, such as core 100b, constructed according to the present disclosure. In this case, core 100b constructed according to the present disclosure may, for example, perform data exchange according to a streaming media protocol with an external unit arranged externally relative to the multi-core system 1000.
[0059] In some embodiments, see Figure 6The multi-core system 1000a has at least one of the following: a) a management core E1, for example, for managing at least one aspect of the multi-core system 1000a; or b) a processing core E2, which has and / or characterizes at least one processing device, such as a computing device, such as a processor device, such as a central processing unit device, such as a CPU; or c) a graphics processing unit core, such as a GPU core E3; or d) an AI core E4 for performing aspects of artificial intelligence, such as performing aspects of machine learning methods, wherein, for example, the AI core E4 has a local storage device E4a, for example, for at least temporarily storing data for performing aspects of artificial intelligence, such as parameters, such as weights or hyperparameters, for at least one artificial neural network NN.
[0060] In some embodiments, see Figure 6 The management chip E5 has multiple processing devices E5a and E5b, such as ISPs.
[0061] In some embodiments, see Figure 6 The management chip E5 has a memory controller E1a for connecting to an external storage device E6.
[0062] In some embodiments, multiple data sources DQ-1, DQ-2 are provided, which, for example, provide video data or image data for processing, for example, by the multi-core system 1000a. For example, reference numeral DQ-1 indicates multiple video data sources, such as cameras, which provide multiple video data streams, for example, four video data streams, each with a resolution of, for example, 3 megapixels and a frame rate of, for example, approximately 44 Hz. For example, reference numeral DQ-2 indicates multiple video data sources, such as cameras, which provide multiple video data streams, for example, five video data streams, each with a resolution of, for example, 8 megapixels and a frame rate of, for example, approximately 35 Hz.
[0063] In some embodiments, see Figure 6 The management chip E1 and the AI chip E4 are configured to exchange data with each other according to a streaming media protocol, for example via a first data connection DV-SP-1 configured to perform a streaming media protocol or a streaming media protocol-based method.
[0064] In some embodiments, see Figure 6 The management chip E1 and the graphics processor chip E3 are configured to exchange data with each other according to a streaming media protocol, for example via a second data connection DV-SP-2 configured to perform a streaming media protocol or a method based on a streaming media protocol.
[0065] In some embodiments, see Figure 6The management core E1 and the processing core E2 are configured to exchange data with each other according to a streaming media protocol, for example via a third data connection DV-SP-3 configured to perform a streaming media protocol or a method based on a streaming media protocol.
[0066] In some embodiments, see Figure 6 At least two of the cores, such as CPU core E2 and GPU core E3, are configured to exchange data with each other, for example directly with each other, for example without using management core E1, according to a streaming media protocol, for example via a fourth data connection DV-SP-4 configured to perform a streaming media protocol or a method based on a streaming media protocol.
[0067] In some embodiments, the above embodiments can be combined arbitrarily, such as any combination of data connections DV-SP-1, DV-SP-2, DV-SP-3, and DV-SP-4.
[0068] For example, the reference numeral "N" in the attached diagram may indicate, for example, multiple parts of the entire dataset at the data connection DV-SP-1, i.e., multiple rows or tiles, such as rows or tiles of a single frame, which are exchanged via the data connection DV-SP-1 according to the streaming protocol. Optionally, at least one additional control line may be provided (see also...). Figure 4 (The attached figure is labeled SL).
[0069] In some embodiments, data connections DV-SP-1, DV-SP-2, DV-SP-3, and DV-SP-4 can each operate using the same streaming media protocol or based on the same streaming media protocol SP.
[0070] In some embodiments, at least some of the data connections DV-SP-1, DV-SP-2, DV-SP-3, and DV-SP-4 may operate, for example, with different streaming media protocols or with different parameters regarding data exchange according to the streaming media protocol.
[0071] In some embodiments, see Figure 6 The memory E6 can at least temporarily store portions N-1, N-2, N-3, ... of data that can be processed or has been processed by the multi-core system 1000a.
[0072] Figure 7 Some embodiments relate to an apparatus 300 for performing the methods according to the present disclosure.
[0073] In other embodiments, see Figure 7The specified device 300 includes: a computing device (“computer”) 302 having at least one computing core 302a, and a storage device 304 associated with the computing device 302, the storage device being used to at least temporarily store at least one of the following: a) data DAT (e.g., at least a portion of data associated with and / or to be transmitted according to the streaming media protocol SP), b) a computer program PRG, for example for performing the method according to the embodiment.
[0074] In other embodiments, see Figure 7 The storage device 304 has volatile memory (e.g., working memory (RAM)) 304a, and / or has non-volatile (NVM) memory (e.g., flash EEPROM) 304b, or a combination thereof or a combination with other memory types not explicitly mentioned.
[0075] Figure 7 Other embodiments relate to a computer-readable storage medium SM containing instructions PRG that, when executed by a computer 302, cause the computer to perform the method according to the embodiments.
[0076] Figure 7 Other embodiments relate to a computer program PRG that includes instructions that, when executed by computer 302, cause the computer to perform the method according to the embodiment.
[0077] Figure 7 Other embodiments involve a data carrier signal DCS, which characterizes and / or transmits a computer program PRG according to the embodiments. The data carrier signal DCS can be received, for example, via an optional data interface 206 of device 200. Similarly, data D can be transmitted (sent and / or received), for example, via the optional data interface 206. In other embodiments, with other units 30 ( Figure 9 Communication, or data communication, can also be conducted through the optional data interface 206.
[0078] In some embodiments, the device 300 or corresponding functions are integrated into at least one core 100, E1 and / or multi-core system.
[0079] Figure 8 Some embodiments relate to products, such as control devices 1, such as control devices for vehicles 10, such as motor vehicles, having at least one of the following: a) a core 100 according to the present disclosure, or b) a core system 1000 according to the present disclosure, or c) a device 300 according to the present disclosure.
[0080] Figure 2Some embodiments relate to a method for a chip 100 for processing data associated with at least one data source, including: exchanging data DAT-DQ with at least one other device 100', such as at least one other chip, according to a streaming media protocol SP.
[0081] Figure 9 Some embodiments relate to using a chip 100 according to the present disclosure and / or a multi-chip system 1000, 1000a according to the present disclosure and / or a product 10 according to the present disclosure and / or a method according to the present disclosure and / or an apparatus 300 according to the present disclosure and / or a computer-readable storage medium SM according to the present disclosure and / or a computer program PRG according to the present disclosure and / or a data carrier signal DCS according to the present disclosure for at least one of the following: a) exchanging 401 data with at least one chip 100, 100', 100a, 100b, E1, E2, E3, E4 according to a streaming media protocol SP; or b) using aspects of the 204 streaming media protocol SP for data exchange within the multi-chip system 1000, 1000a; c) using aspects of the 403 streaming media protocol SP for data exchange between at least one component of the multi-chip system 1000, 1000a and at least one component or device located external to the multi-chip system 1000, 1000a; or d) Reduce 404 latency, such as latency in processing and / or playing, for example, image data and / or video data, or e) reduce 405 the storage bandwidth requirements for at least one interface with respect to the chip and / or for at least one storage device E6 for the chip and / or for a multi-chip system.
Claims
1. A core (100) for processing data (DAT-DQ) associated with at least one data source (DQ; DQ'; DQ''; DQ'''), wherein, The core (100) is configured to exchange (200) the data (DAT-DQ) with at least one other device (100'; DS), such as at least one other core (100'), according to the streaming media protocol (SP).
2. The core (100) according to claim 1, wherein, The data (DAT-DQ) has at least one of the following: a) image data (DAT-BILD), or b) video data (DAT-VID).
3. The core (100) according to at least one of the preceding claims, wherein, The exchange (200) performed according to the streaming media protocol (SP) has at least one of the following: a) exchanging the data (DAT-DQ) said in (200a) using, for example, an existing, such as a standardized streaming media protocol (SP), or b) exchanging the data (DAT-DQ) said in (200b) based on the streaming media protocol (SP).
4. The core (100) according to at least one of the preceding claims, wherein, The core (100) is configured to process (202) the data (DAT-DQ).
5. The core (100) according to claim 4, wherein, The process (202) has at least one of the following: a) processing (202a) a portion (DAT-DQ-TEIL) of the data (DAT-DQ), for example before, for example, fully receiving the data (DAT-DQ) and / or during the transmission of the data (DAT-DQ), wherein, for example in the case of image data (DAT-BILD), the portion (DAT-DQ-TEIL) of the data (DAT-DQ) represents and / or has at least one of the following: a1) a row, or a2) an image element, such as a pixel, or a3) a tile.
6. The core (100) according to at least one of the preceding claims, wherein, The core (100) is configured to perform at least one of the following: a) receiving (210) the data (DAT-DQ) from the at least one other device, such as the at least one other core (100'), or b) receiving (212) the data (DAT-DQ) from at least one data source (DQ; DQ'; DQ''; DQ'''), or c) transmitting (214) the data (DAT-DQ) to the at least one other device (100'), such as the at least one other core (100').
7. The chip (100) according to at least one of the preceding claims has a memory (110), such as a buffer memory, for at least temporarily storing at least a portion of the data (DAT-DQ).
8. The chip (100) according to at least one of the preceding claims has a communication system (120), such as an on-chip network, such as a cross switch, for exchanging data within the chip (100), such as exchanging data with a processor or computing core (130), and / or exchanging data with at least one external device (140), such as an external storage device.
9. The chip (100) according to at least one of the preceding claims has at least one processing unit (E5a, E5b) configured for processing image data (DAT-BILD) and / or video data (DAT-VID), for example dedicated to processing image data (DAT-BILD) and / or video data (DAT-VID), for example having at least one image signal processor.
10. A multi-core system (1000; 1000a) having multiple cores (100a, 100b; E1, E2, E3, E4), wherein, At least one of the plurality of cores (100a, 100b; E1, E2, E3, E4) is constructed according to at least one of the preceding claims.
11. The multi-core system (1000; 1000a) according to claim 10, comprising at least one of the following: a) a management core (E1), for example for managing at least one aspect of the multi-core system (1000; 1000a), or b) a processing core (E2), the processing core having and / or characterizing at least one processing device, such as a computing device, such as a processor device, such as a central processing unit device, such as a CPU, or c) a graphics processing core, such as a GPU core, or d) an AI core (E4) for performing aspects of artificial intelligence, such as performing aspects of machine learning methods, wherein, For example, the AI chip (E4) has a local storage device (E4a) for storing, for example, data for at least temporarily performing various aspects of artificial intelligence, such as parameters, weights, or hyperparameters, for at least one artificial neural network (NN).
12. The multi-core system (1000; 1000a) according to claim 11, wherein, a) The management chip (E1) and the AI chip (E4) are configured to exchange data with each other according to a streaming media protocol (SP), for example via a first data connection (DV-SP-1) configured to perform a method of the streaming media protocol (SP) or based on the streaming media protocol (SP), or wherein, b) The management chip (E1) and the graphics processor chip (E3) are configured to exchange data with each other according to a streaming media protocol (SP), for example via a second data connection (DV-SP-2) configured to perform a method of the streaming media protocol (SP) or based on the streaming media protocol (SP), or wherein, c) The management chip (E1) and the processing chip (E2) are configured to exchange data with each other according to a streaming media protocol (SP), for example via a third data connection (DV-SP-3) configured to perform a method of the streaming media protocol (SP) or based on the streaming media protocol (SP), or wherein, d) At least two of the cores (E2, E3, E4) are configured to exchange data with each other, for example directly with each other, for example without using the management core (E1), according to the streaming media protocol (SP), for example via a fourth data connection (DV-SP-4) configured to perform the streaming media protocol (SP) or a method based on the streaming media protocol (SP).
13. A product (1), such as a control device, for example for a vehicle (10), such as a motor vehicle, said product having at least one of the following: a) a core (100) according to at least one of the preceding claims 1 to 9, or b) a core system (1000; 1000a) according to at least one of the preceding claims 10 to 12, or c) a device (300) according to claim 12.
14. A method for use with a core (100) for processing with at least one data source (DQ; DQ'; Data associated with DQ''; DQ''') (DAT-DQ), including: data associated with at least one other device (100'; according to the streaming media protocol (SP); DS), for example at least one other core (100'), exchange (200) data.
15. An apparatus (300) for performing the method according to claim 14.
16. A computer-readable storage medium (SM) comprising instructions (PRG) that, when executed by a computer (302), cause the computer to perform the method according to claim 14.
17. A computer program (PRG) comprising instructions that, when executed by a computer (302), cause the computer to perform the method according to claim 14.
18. A data carrier signal (DCS) that transmits and / or characterizes the computer program (PRG) according to claim 17.
19. An application (400) of a core (100) according to at least one of claims 1 to 9 and / or a multi-core system (1000) according to at least one of claims 10 to 12 and / or a product (1) according to claim 13 and / or a method according to claim 14 and / or an apparatus (300) according to claim 15 and / or a computer-readable storage medium (SM) according to claim 16 and / or a computer program (PRG) according to claim 17 and / or a data carrier signal (DCS) according to claim 18, for at least one of the following: a) exchanging (401) data with at least one core (100) according to a streaming media protocol (SP), or b) using (402) aspects of the streaming media protocol (SP) for data exchange within the multi-core system (1000; 1000a), c) Using (403) aspects of the streaming media protocol (SP) for data exchange between at least one component (100; E1) of the multi-core system (1000; 1000a) and at least one component or device located outside the multi-core system (1000; 1000a), or d) reducing (404) latency, for example, reducing latency when processing and / or playing, for example, image data and / or video data, or e) reducing (405) the storage bandwidth requirements for at least one interface of the core (100) and / or for at least one storage device for the core (100) and / or for the multi-core system (1000; 1000a), f) data processing in the field of motor vehicles, such as continuous data processing, such as image data processing, such as video data processing.