NVIDIA had 31 patents in cloud during Q4 2023. NVIDIA Corp filed patents in Q4 2023 for methods to detect inefficiencies in workloads, transferring large files over a network efficiently using QUIC and HTTP/3, training language models on structural database relationships, implementing a combined on-package and off-package memory system for improved bandwidth and capacity, and identifying buffers for post-processing applications based on buffer statistics. GlobalData’s report on NVIDIA gives a 360-degreee view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
NVIDIA grant share with cloud as a theme is 38% in Q4 2023. Grant share is based on the ratio of number of grants to total number of patents.
Recent Patents
Application: Detecting and optimizing program workload inefficiencies at runtime (Patent ID: US20230418726A1)
The patent filed by NVIDIA Corp. describes methods and systems for identifying potential inefficiencies in workloads by comparing information obtained during execution to a set of inefficiency patterns. The system can detect inefficiencies based on metrics of hardware platform utilization, calls to operations, values generated by operations, and parameters included in calls. It can also automatically determine actions to address inefficiencies, use machine learning to optimize actions, and provide users with potential improvements achievable by implementing these actions. Additionally, the system can generate modified workloads to test improvements and prioritize actions based on the severity of inefficiencies detected.
Furthermore, the patent outlines a system with circuits that automatically select workloads, compare information to inefficiency patterns, and identify potential inefficiencies. These circuits can determine actions to address inefficiencies, provide these actions to users, and measure performance improvements. The system can also automatically select workloads from previous executions, enter issues into tracking systems, and operate in various environments such as autonomous machines, simulation systems, deep learning operations, and cloud computing resources. Additionally, the patent describes a processor with neural networks that can identify potential inefficiencies in workload execution based on information obtained and predefined inefficiency patterns. The processor can import performance analyzer instructions into workloads, patch instructions for operations, and conduct static analyses or telemetry measurements to optimize performance and automatically implement actions to improve inefficiencies.
To know more about GlobalData’s detailed insights on NVIDIA, buy the report here.
Premium Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
