IDiyas

For the Inventor, By the Inventor

Austin, TX, United States of America

Omkar Paranjape

USPTO Granted Patents = 16 

Average Co-Inventor Count = 9.6

ph-index = 3

Forward Citations = 34(Granted Patents)


Company Filing History:

goldMedal16 out of 5,406 
 
Nvidia Corporation
 patents

Years Active: 2015-2024

Loading Chart...
16 patents (USPTO):
1. 11907717 - Techniques for efficiently transferring data to a processor (2024-02-20),
2. 11768686 - Out of order memory request tracking structure and technique (2023-09-26),
3. 11604649 - Techniques for efficiently transferring data to a processor (2023-03-14),
4. 11379944 - Techniques for performing accelerated point sampling in a texture processing pipeline (2022-07-05),
5. 11367160 - Simultaneous compute and graphics scheduling (2022-06-21),
6. 11347668 - Unified cache for diverse memory traffic (2022-05-31),
7. 11080051 - Techniques for efficiently transferring data to a processor (2021-08-03),
8. 10705994 - Unified cache for diverse memory traffic (2020-07-07),
9. 10459861 - Unified cache for diverse memory traffic (2019-10-29),
10. 10152329 - Pre-scheduled replays of divergent operations (2018-12-11),
11. 10095548 - Mechanism for waking common resource requests within a resource management subsystem (2018-10-09),
12. 9836325 - Resource management subsystem that maintains fairness and order (2017-12-05),
13. 9817668 - Batched replays of divergent operations (2017-11-14),
14. 9755994 - Mechanism for tracking age of common resource requests within a resource management subsystem (2017-09-05),
15. 9626191 - Shaped register file reads (2017-04-18),

Title: Innovations of Omkar Paranjape

Introduction

Omkar Paranjape is a notable inventor based in Austin, TX, with a remarkable portfolio of 16 patents. His work primarily focuses on enhancing data transfer techniques and improving memory request tracking in multiprocessor systems.

Latest Patents

One of his latest patents is titled "Techniques for efficiently transferring data to a processor." This invention discloses a method for block data transfer that minimizes data transfer and memory access overheads. It significantly reduces multiprocessor activity and energy consumption. Threads executing on a multiprocessor can request and store needed data in on-chip shared memory, allowing for multiple accesses. The data is loaded from global memory into shared memory using an instruction that bypasses the need for storing data in registers or cache memory during the transfer. Another significant patent is the "Out of order memory request tracking structure and technique." This invention employs multiple, dynamically sized tracking queues in a streaming cache. It distributes request tracking information among these queues to enable selective out-of-order memory request returns. A dynamically controlled policy assigns pending requests to tracking queues, allowing for in-order memory returns in some contexts while enabling out-of-order returns in others.

Career Highlights

Omkar Paranjape currently works at Nvidia Corporation, a leading technology company known for its innovations in graphics processing and computing. His contributions to the field have been instrumental in advancing data processing techniques.

Collaborations

He collaborates with talented coworkers such as Shirish C Gadre and Steven James Heinrich, who contribute to the innovative environment at Nvidia Corporation.

Conclusion

Omkar Paranjape's contributions to data transfer and memory request tracking have significantly impacted the field of computer engineering. His innovative patents reflect his dedication to improving technology and efficiency in multiprocessor systems.

This text is generated by artificial intelligence and may not be accurate.
Please report any incorrect information to support@idiyas.com
Loading…