PATENT

Patent Abstract

SLA-Constrained Energy-Aware Inference Scheduling Using Memory Residency, DMA Transfer Policy, Variant Selection, and Performance-State Control on ARM Edge Systems

Patent application 202641005501 in the Manish KL patent portfolio, covering edge ai scheduling and related technical systems.

202641005501 2026-01-20 Edge AI Scheduling
Full Patent Record India Filings Table

Summary

Overview

This invention targets edge AI systems that must minimize energy per inference while still satisfying latency constraints.

Abstract

Technical Abstract

A controller running on an ARM-based system predicts latency and energy for candidate execution policies and selects the best one under current request and hardware conditions. Policy dimensions include SRAM residency, no-waste DMA prefetch, precompiled graph-variant selection, and performance-state choices such as DVFS, with runtime telemetry continuously improving the predictive models.

Search Context

SEO Keywords

edge AI patent, ARM inference scheduling patent, energy aware inference patent, DMA policy patent, latency constrained inference patent

Related Patents

More Patents in accelerator architecture, compiler control, and cluster systems

These filings sit nearby in the portfolio and strengthen internal linking across related patent topics.

Accelerator Architecture

SRMIC-X1: SRAM-Residency Memory-Centric Inference Chip - A Residency-First LLM Decode Accelerator Architecture

SRMIC-X1 is a residency-first inference accelerator that treats the active per-token working set as a hardware primitive and prioritizes hot SRAM residency over HBM-bound decode behavior.

 Kernel Autotuning

Systems and Methods for Noise-Aware Benchmarking and Optimization of AI-Generated Compute Kernels

This filing defines a reliability-gated benchmarking loop for generated accelerator kernels so noisy measurements do not drive bad kernel-selection decisions.

 Compiler and I/O Scheduling

Unified Storage-Aware Accelerator Compilation with Deterministic I/O Windows and Bounded Stall Guarantees

Storage and network movement are elevated into accelerator compilation so compute windows and data windows can be aligned with deterministic slack and bounded stall behavior.