Compute optimization mechanism for deep neural networks

US10417731B2 · kind B2 · utility

Assignee

• Intel Corporation · US

Inventors

• Prasoonkumar Surti · Folsom, US
• Narayan Srinivasa · Thousand Oaks, US
• Feng Chen · Shanghai, CN
• Joydeep Ray · Folsom, US
• Ben J. Ashbaugh · Folsom, US
• Nicolas C. Galoppo Von Borries · Portland, US
• Eriko Nurvitadhi · Beaverton, US
• Balaji Vembu · Folsom, US
• Tsung-Han Lin · Campbell, US
• Kamal Sinha · Rancho Cordova, US
• Rajkishore Barik · Santa Clara, US
• Sara S. Baghsorkhi · San Jose, US
• Justin E. Gottschlich · Santa Clara, US
• Altug Koker · Cameron Park, US
• Nadathur Rajagopalan Satish · Santa Clara, US
• Farshad Akhbari · Chandler, US
• Dukhwan Kim · Mountain View, US
• Wenyin Fu · Folsom, US
• Travis T. Schluessler · Hillsboro, US
• Josh B. Mastronarde · Sacramento, US
• Linda L. Hurd · Auburn Lake Trails, US
• John H. Feit · Folsom, US
• Jeffery S. Boles · Folsom, US
• Adam T. Lake · Portland, US
• Karthik Vaidyanathan · San Francisco, US
• Devan Burke · Portland, US
• Subramaniam Maiyuran · Citrus, US
• Abhishek R. Appu · Cameron Park, US

Key dates

• Filing date: Apr 24, 2017
• Grant date: Sep 17, 2019
• Expiry date: Apr 24, 2037

Classification

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

Abstract

An apparatus to facilitate compute optimization is disclosed. The apparatus includes a plurality of processing units each comprising a plurality of execution units (EUs), wherein the plurality of EUs comprise a first EU type and a second EU type.