	US 11,941,007 B2
	Method for reducing the computing time of a data processing unit
Daniel Jansen, Friedrichsfehn (DE)
Assigned to ODASS GBR, Hamburg (DE)
Appl. No. 16/956,482
Filed by ODASS GbR, Hamburg (DE)
PCT Filed Dec. 19, 2018, PCT No. PCT/EP2018/085803 § 371(c)(1), (2) Date Jun. 19, 2020, PCT Pub. No. WO2019/121890, PCT Pub. Date Jun. 27, 2019.
Claims priority of application No. 10 2017 131 142.4 (DE), filed on Dec. 22, 2017; and application No. 10 2018 126 546.8 (DE), filed on Oct. 24, 2018.
Prior Publication US 2020/0349167 A1, Nov. 5, 2020
Int. Cl. G06F 16/2455 (2019.01); G06F 16/22 (2019.01); G06F 16/28 (2019.01)

CPC G06F 16/24568 (2019.01) [G06F 16/2255 (2019.01); G06F 16/285 (2019.01)]

18 Claims

1. A method for reducing the computing time of a data processing unit during processing of data,

wherein

an input data stream consisting of data stream elements is read and detected sequentially by at least one data processing unit,

each data stream element of the input data stream is divided up into element components,

the element components are grouped into component groups in accordance with different categories,

at least individual component groups are checked for coincidence to determine whether the component groups have already been detected, wherein component groups which do not coincide with the already detected component groups comprise new content to the already detected component groups,

at least one element-addressable secondary data stream is generated, by the data processing unit, from the component groups not coinciding with already detected component groups,

wherein the element-addressable secondary data stream contains only the component groups not coincide with the already detected component groups in the element-addressable secondary data stream,

wherein the element addressability of the secondary data stream is established by the assignment of an index data stream to the secondary data stream,

wherein an address of each component group of the secondary data stream is stored in the index data stream,

wherein the index data stream includes slots each containing index data and corresponding element components, the index data each indicating an address of one of the component groups in the element-addressable secondary data stream,

wherein the index data being of the same length among each other, wherein the length is the number of element components of an index datum, thus that a position in the index data stream is calculable due to the same length of the index data by multiplying this length of the index data with the position of the index datum to be calculated, wherein a position of an n-th index datum can be calculated by multiplying n with the length of the index data,

wherein a position of a secondary data stream element in the secondary data stream can be determined, thereby that the number of element components of the secondary data stream components located in the secondary data stream prior to the secondary data stream element to be determined is contained in the index datum assigned to the secondary data stream element,

wherein an order of index data corresponds to an order of allocated assigned data stream elements in the secondary data stream,

an output data stream is generated by inputting the slots in the element-addressable secondary data stream,

wherein the output data stream contains the element components in the input data stream and slot positions of the index data stream, each slot position containing the address of the corresponding element component in the secondary data stream,

wherein a sequence of the address of the data stream elements of the output data stream corresponds to the sequence of data stream elements of the input data stream,

wherein the input data stream is completely restorable from the output data stream with the same sequence of data elements via the element addresses of the secondary data stream included in the index data stream, and

the output data stream is fed to a further processing in the data processing unit.