This inventor holds 1 USPTO granted patent. Top assignee: Oracle America, Inc.. Active years: 2014.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2014
Title: Innovations by Hong-Men Su in Thread Scheduling Technology
Introduction
Hong-Men Su is an accomplished inventor based in San Mateo, CA (US). He has made significant contributions to the field of microprocessor technology, particularly in the area of multithreading. His innovative approach to scheduling multiple threads has garnered attention in the tech community.
Latest Patents
Hong-Men Su holds a patent for a "Method and apparatus for scheduling multiple threads for execution in a shared microprocessor pipeline." This patent describes an apparatus and method for scheduling the execution of multiple threads on a shared processor resource in connection with a multithreaded multiprocessor chip. The invention employs a thread selection policy that switches between available threads every cycle, prioritizing the least recently executed or scheduled threads. This approach ensures that different threads can operate without encountering deadlocks or livelocks, while maximizing aggregate performance and fairness among threads. The prioritization is achieved by monitoring and sorting thread status information, including speculative states, which enhances the usage of the execution pipeline.
Career Highlights
Hong-Men Su is currently employed at Oracle America, Inc., where he continues to develop innovative solutions in the field of computing. His work has had a lasting impact on the efficiency and performance of multithreaded processing systems.
Collaborations
One of his notable coworkers is Kathirgamar Aingaran, with whom he has likely collaborated on various projects within Oracle America, Inc.
Conclusion
Hong-Men Su's contributions to thread scheduling technology exemplify the innovative spirit that drives advancements in microprocessor design. His patent reflects a deep understanding of multithreading and its challenges, paving the way for more efficient computing solutions.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
