Andrew Kinoshita Petersen

Title: Andrew Kinoshita Petersen: Innovator in Microarchitecture

Introduction

Andrew Kinoshita Petersen is a notable inventor based in Seattle, WA (US). He has made significant contributions to the field of microarchitecture, particularly through his innovative patent related to wavecache technology. His work focuses on enhancing the efficiency of executing multiple threads in computing systems.

Latest Patents

Andrew holds a patent for "Building a wavecache," which introduces a microarchitecture and instruction set that supports multiple, simultaneously executing threads. This approach is particularly relevant to a microarchitecture known as 'WaveScalar.' WaveScalar is a compiler that segments a control flow graph into pieces called waves, which contain partially ordered instructions. This design allows for control to enter at a single point, thereby optimizing execution. The patent also discusses the provision of instructions that enable and disable wave-ordered memory, as well as additional memory access instructions that bypass wave-ordered memory to expose further parallelism. Furthermore, a lightweight interthread synchronization model is employed, which mimics hardware queue locks, and a simple fence instruction is introduced to manage relaxed memory consistency.

Career Highlights

Andrew is affiliated with the University of Washington, where he continues to contribute to advancements in microarchitecture. His work has implications for both specialized and conventional microarchitectures, making it relevant across various computing applications.

Collaborations

Andrew has collaborated with esteemed colleagues such as Susan Jane Eggers and Steven James Swanson. Their collective expertise enhances the research and development efforts in the field of microarchitecture.

Conclusion

Andrew Kinoshita Petersen's innovative work in microarchitecture, particularly through his patent on wavecache technology, showcases his commitment to advancing computing efficiency. His contributions are significant in the realm of executing multiple threads, and his collaborations further enrich the field.