Compression of analog content addressable memory
AssigneeHewlett Packard Enterprise Development LP
InventorsGiacomo Pedretti, Catherine Graves, John Paul Strachan
Life of this patent
Estimating the term…
Abstract
A technique for compressing an analog content addressable memory (CAM) array is provided. Random input data is applied to the analog CAM array, and an average measure of similarity is calculated for each output row of the analog CAM array. Rows of the analog CAM array that have measures of similarity that are close to each other can be eliminated, such as by removing similar rows or merging together similar rows. Thus, the analog CAM array size can be reduced without a loss in accuracy of a model stored on the analog CAM array.
Claims20
A content addressable memory (CAM) device comprising:
- an analog CAM array, a first row of the analog CAM array comprising:
- first analog CAM cells, each of the first analog CAM cells storing data that is associated with a respective first stored range of analog values having a first lower limit and a first upper limit;
- and
- a first match line horizontally aligned with and coupled to the first analog CAM cells;
- and
- a first discharge line horizontally aligned with and coupled to the first analog CAM cells, wherein each of the first analog CAM cells is configured to contribute to a first discharge current on the first discharge line based on whether a respective search value is within the respective first stored range of analog values;
- a first transimpedance amplifier configured to be coupled to the first discharge line, the first transimpedance amplifier configured to convert the first discharge current into a first analog output;
- and
- a first analog-to-digital converter coupled to the first transimpedance amplifier, the first analog-to-digital converter configured to convert the first analog output to a first digital output.
A content addressable memory (CAM) device comprising:
- an analog CAM array, a first row of the analog CAM array comprising:
- a first analog CAM cell, the first analog CAM cell comprising:
- a first memristor storing a first analog value;
- and
- a second memristor storing a second analog value;
- a first match line coupled to the first analog CAM cell;
- and
- a first discharge line coupled to the first analog CAM cell, wherein the first analog CAM cell is configured to contribute to a first discharge current on the first discharge line based on whether a search value is between the first analog value and the second analog value;
- and
- a sensing circuit comprising:
- a first transimpedance amplifier configured to be coupled to the first discharge line, the first transimpedance amplifier configured to convert the first discharge current into a first analog output;
- and
- a first analog-to-digital converter coupled to the first transimpedance amplifier, wherein the first analog-to-digital converter is configured to generate a first digital representation of the first analog output.
A content addressable memory (CAM) compression method comprising:
- programming a range of first voltages into first analog CAM cells of a first row of an analog CAM array;
- programming a range of second voltages into second analog CAM cells of a second row of the analog CAM array;
- applying an input vector comprising a set of random inputs to the first analog CAM cells, wherein each of the first analog CAM cells is configured to contribute to a first discharge current on a first discharge line that is coupled to the first analog CAM cells based on whether a respective random input applied to a respective first analog CAM cell is within the range of the first voltages of the respective first analog CAM cell;
- applying the input vector to the second analog CAM cells, wherein each of the second analog CAM cells is configured to contribute to a second discharge current on a second discharge line that is coupled to the second analog CAM cells based on whether a respective random input applied to a respective second analog CAM cell is within the range of the second voltages of the respective second analog CAM cell;
- converting the first discharge current to a first digital representation that represents a first measure of similarity of the first row of the analog CAM array to the set of random inputs;
- and
- converting the second discharge current to a second digital representation that represents a second measure of similarity of the second row of the analog CAM array to the set of random inputs.
Dependencies are inferred from the claim text.