Graphics processing unit processing and caching improvements

US11861761B2 · kind B2 · utility

Assignee

• Intel Corporation · US

Inventors

• Subramaniam Maiyuran · Citrus, US
• Durgaprasad Bilagi · Folsom, US
• Joydeep Ray · Folsom, US
• Scott Janus · Rocklin, US
• Sanjeev Jahagirdar · Folsom, US
• Brent Insko · Portland, US
• Lidong Xu · Beijing, CN
• Abhishek R. Appu · Cameron Park, US
• James M. Holland · Folsom, US
• Vasanth Ranganathan · Cameron Park, US
• Nikos Kaburlasos · Lincoln, US
• Altug Koker · Cameron Park, US
• Xinmin Tian · Union City, US
• Guei-Yuan Lueh · San Jose, US
• Changliang Wang · Bellevue, US

Key dates

• Filing date: Nov 11, 2020
• Grant date: Jan 2, 2024
• Expiry date: Feb 12, 2041

Classification

• Technology area (CPC Y): Emerging Cross-Sectional Technologies
• CPC primary: Y02D10/00
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Embodiments described herein are generally directed to improvements relating to power, latency, bandwidth and/or performance issues relating to GPU processing/caching. According to one embodiment, a system includes a producer intellectual property (IP) (e.g., a media IP), a compute core (e.g., a GPU or an AI-specific core of the GPU), a streaming buffer logically interposed between the producer IP and the compute core. The producer IP is operable to consume data from memory and output results to the streaming buffer. The compute core is operable to perform AI inference processing based on data consumed from the streaming buffer and output AI inference processing results to the memory.