# US 2025/0166115 A1 · Compute optimization mechanism for deep neural networks

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0166115 A1
- Document ID: US20250166115A1
- Status: Published application
- Kind: A1
- Application number: 18/971,949
- Filed: 2024-12-06
- Published: 2025-05-22
- Earliest claimed: 2017-04-24
- Assignees: Intel Corporation
- Applicants: Intel Corporation
- Inventors: Prasoonkumar Surti; Narayan Srinivasa; Feng Chen; Joydeep Ray; Ben J. Ashbaugh; Nicolas C. Galoppo Von Borries; Eriko Nurvitadhi; Balaji Vembu; Tsung-Han Lin; Kamal Sinha; Rajkishore Barik; Sara S. Baghsorkhi; Justin E. Gottschlich; Altug Koker; Nadathur Rajagopalan Satish; Farshad Akhbari; Dukhwan Kim; Wenyin Fu; Travis T. Schluessler; Josh B. Mastronarde; Linda L. Hurd; John H. Feit; Jeffery S. Boles; Adam T. Lake; Karthik Vaidyanathan; Devan Burke; Subramaniam Maiyuran; Abhishek R. Appu
- Web page: https://patentgrep.com/patent/US20250166115A1

## CPC

- G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining
- G06F 9/45533: {Hypervisors; Virtual machine monitors}
- G06F 9/5061: {Partitioning or combining of resources}
- G06F 9/5094: {where the allocation takes into account power or heat criteria (power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206)}
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/063: using electronic means
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N 3/0895: Weakly supervised learning, e.g. semi-supervised or self-supervised learning
- G06N 3/09: Supervised learning
- G06N 3/098: Distributed learning, e.g. federated learning
- G06F 8/41: Compilation
- G06F 2009/45583: {Memory management, e.g. access or allocation}
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

Embodiments provide mechanisms to facilitate compute operations for deep neural networks. One embodiment comprises a graphics processing unit comprising one or more multiprocessors, at least one of the one or more multiprocessors including a register file to store a plurality of different types of operands and a plurality of processing cores. The plurality of processing cores includes a first set of processing cores of a first type and a second set of processing cores of a second type. The first set of processing cores are associated with a first memory channel and the second set of processing cores are associated with a second memory channel.

## Claims

### Claim 1

A graphics processing apparatus comprising:
an interconnect to a host processor; and
a plurality of graphics processing clusters respectively including a plurality of multiprocessors coupled via an interconnect, a graphics processing cluster of the plurality of graphics processing clusters including:
a register file to store a plurality of different types of operands;
a first plurality of processing resources of a first type configurable to process a first number of threads having operands stored in a first number of registers of the register file; and
a second plurality of processing resources of a second type configurable to process a second number of threads having operands stored in a second number of registers of the register file, the first number of threads greater than the second number of threads and the second number of registers greater than the first number of registers.

### Claim 2 (depends on 1)

The graphics processing apparatus of claim 1, wherein the first plurality of processing resources is configured to perform multi-dimensional matrix operations on the operands stored in the first number of registers.

### Claim 3 (depends on 1)

The graphics processing apparatus of claim 1, wherein the second plurality of processing resources is configured to perform graphics operations on the operands stored in the second number of registers.

### Claim 4 (depends on 1)

The graphics processing apparatus of claim 1, further comprising compute circuitry to select processing resources to execute a workload.

### Claim 5 (depends on 4)

The graphics processing apparatus of claim 4, wherein the compute circuitry is to select processing resources of the first type to process a first type of application workload and to select the processing resources of the second type to process a second type of application workload.

### Claim 6 (depends on 1)

The graphics processing apparatus of claim 1, further comprising a memory device coupled with the plurality of graphics processing clusters.

### Claim 7 (depends on 6)

The graphics processing apparatus of claim 6, wherein the memory device includes a high bandwidth memory (HBM) including a plurality of memory channels.

### Claim 8 (depends on 7)

The graphics processing apparatus of claim 7, wherein a first memory channel of the HBM is configured to couple with one or more processing resources of the first plurality of processing resources of a first type and a second memory channel of the HBM is configured to couple with one or more processing resources of the second plurality of processing resources of a second type.

### Claim 9 (depends on 1)

The graphics processing apparatus of claim 1, wherein the register file is configured to perform matrix-vector transformations.

### Claim 10 (depends on 1)

The graphics processing apparatus of claim 1, further comprising a shared local memory (SLM) configured to perform matrix-vector transformations.

### Claim 11

A method comprising:
storing operands for a plurality of different types of operations to a register file of a graphics processing including a plurality of graphics processing clusters of a graphics processor, each of the plurality of graphics processing clusters including a plurality of multiprocessors coupled via an interconnect;
processing a first number of threads having operands stored in a first number of registers of the register file via a first plurality of processing resources of a first type; and
processing second number of threads having operands stored in a second number of registers of the register file via a second plurality of processing resources of a second type, the first number of threads greater than the second number of threads and the second number of registers greater than the first number of registers.

### Claim 12 (depends on 11)

The method of claim 11, comprising:
performing multi-dimensional matrix operations on the operands stored in the first number of registers via the first plurality of processing resources; and
performing graphics operations on the operands stored in the second number of registers via the second plurality of processing resources.

### Claim 13 (depends on 11)

The method of claim 11, further comprising selecting processing resources to execute a workload via compute circuitry of the graphics processor, including selecting processing resources of the first type to process a first type of application workload and selecting the processing resources of the second type to process a second type of application workload.

### Claim 14 (depends on 11)

The method of claim 11, further comprising performing matrix-vector transformations via the register file of the graphics processor.

### Claim 15 (depends on 11)

The method of claim 11, further comprising performing matrix-vector transformations via shared local memory (SLM) of the graphics processor.

### Claim 16

A graphics processing system comprising:
an interconnect to a host processor;
a memory device coupled with the interconnect; and
a graphics processor coupled with the interconnect and the memory device, the graphics processor including a plurality of graphics processing clusters, each of the plurality of graphics processing clusters including a plurality of multiprocessors coupled via an interconnect, a graphics processing cluster of the plurality of graphics processing clusters including:
a register file to store a plurality of different types of operands;
a first plurality of processing resources of a first type configurable to process a first number of threads having operands stored in a first number of registers of the register file; and
a second plurality of processing resources of a second type configurable to process a second number of threads having operands stored in a second number of registers of the register file, the first number of threads greater than the second number of threads and the second number of registers greater than the first number of registers.

### Claim 17 (depends on 16)

The graphics processing system of claim 16, wherein the first plurality of processing resources is configured to perform multi-dimensional matrix operations on the operands stored in the first number of registers and the second plurality of processing resources is configured to perform graphics operations on the operands stored in the second number of registers.

### Claim 18 (depends on 16)

The graphics processing system of claim 16, further comprising compute circuitry to select processing resources to execute a workload, the compute circuitry to select processing resources of the first type to process a first type of application workload and to select the processing resources of the second type to process a second type of application workload.

### Claim 19 (depends on 16)

The graphics processing system of claim 16, wherein the memory device includes a high bandwidth memory (HBM) including a plurality of memory channels, a first memory channel of the HBM is configured to couple with one or more processing resources of the first plurality of processing resources of a first type, and a second memory channel of the HBM is configured to couple with one or more processing resources of the second plurality of processing resources of a second type

### Claim 20 (depends on 16)

The graphics processing system of claim 16, wherein the register file is configured to perform matrix-vector transformations and further comprising a shared local memory (SLM) configured to perform matrix-vector transformations.
