MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085267 A) filed by V. Prema Tulasi; and Dr. A. Raji Reddy on July 12, 2026, for System And Method For Distributed Runtime Optimization Of Energy-Aware Artificial Intelligence Workloads Using Predictive Tensor Scheduling.
Inventors include M. Sivajyothi; Dr. P. Swetha; Dr. K. Shahu Chatrapati; Dr. K. Srujan Raju; Dr. D. Maneiah; and K. Bhargavai Triveni Nandana.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: ABSTRACT A system (100) and method for distributed runtime optimization of energy-aware artificial intelligence workloads using predictive tensor scheduling are disclosed. The system (100) comprises a memory (102), a processor (104), and a communication module (106) configured to communicate with distributed computing nodes and heterogeneous accelerator resources. A workload analysis module (108) generates tensor dependency representations associated with artificial intelligence workloads, while a resource profiling module (110) monitors processor utilization, communication latency, thermal conditions, and energy consumption metrics. A predictive tensor scheduling module (112) generates predictive execution models using runtime telemetry information and historical tensor execution traces to dynamically allocate tensor operations across distributed computing resources. An adaptive orchestration module (114) modifies tensor execution placement during runtime, and an energy optimization module (116) minimizes cumulative runtime energy consumption. Fig. 1
Disclaimer: Curated by HT Syndication.
