Nvidia patents method to speed up math by skipping empty data blocks
Nvidia describes a way to handle large tables of numbers, called sparse matrices, by breaking them into smaller blocks. The system ignores blocks that contain only zeros and only processes the blocks with actual data, using specific labels to keep the numbers aligned. This allows the computer to perform calculations faster and with less energy.
In the filing’s own words
Approaches in accordance with various embodiments provide for the processing of sparse matrices for mathematical and programmatic operations. In particular, various embodiments utilize a tiling approach that divides a sparse matrix into submatrices, many of which will include only zero-value entities. These empty tiles can be ignored, and only the tiles with non-zero entries processed, which reduces resource and time requirements for the processing. An indexing approach can be used for each entity that is a combination of the tile identifier and an offset value, which enables the values to be multiplied correctly against, for example, values of a dense matrix. The tiles can be processed in parallel and the results accumulated to generate a matrix product. The matrix product can then be passed to the next step in a process or operation, such as to a next layer in a deep neural network.
What exactly they claim (claim 1)
1. One or more processors, comprising: circuitry to: perform one or more matrix operations on one or more matrix operands indicated by one or more indices, wherein the one or more indices indicate one or more locations, within a memory resource accessible to the one or more processors, of one or more separately compressed multi-element portions of the one or more matrix operands comprising one or more non-zero values.
Facts
- Company
- Nvidia (NVIDIA Corporation)
- Official title
- TILED COMPRESSED SPARSE MATRIX FORMAT
- Publication number
- US20260267645A1
- Application number
- 19659539
- Filed
- Apr 27, 2026
- Published
- Sep 10, 2026
- Inventors
- 3
- Claims
- 1
Classified under
G06F 9/30036 Physics & computingG06F 9/3877 Physics & computingG06F 9/3885 Physics & computingG06F 17/16 Physics & computingG06N 3/0418 Physics & computingMore from Nvidia in Chips & hardware
- Nvidia patents method to skip zero data in neural network math Oct 13, 2026
- Nvidia patents hardware to speed up motion blur in ray tracing Oct 13, 2026
- Nvidia patents method to sync data across split processor chips Sep 17, 2026
- Nvidia patents chip math trick to speed up AI calculations Oct 8, 2026
- Nvidia patents data center cooling that uses cryogenic fluids Oct 15, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.