Heterogeneous Task Processing With Time-Shifted Output Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Real-time processing of sequences of tasks is limited by the longest computation time, and distributed or heterogeneous task computation introduces latency due to data transportation and cross-core connectivity.
Innovation Solution
A distributed and/or heterogeneous computation architecture that selects between tasks meeting real-time constraints and those that do not, using a spatiotemporal feature transitivity (STFT) function to adjust the output of longer-running tasks to meet the real-time requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distributed or heterogeneous task computation is used to process data blocks, then processing capability is improved, but latency increases due to data transportation and cross-core connectivity
Solution Approach 1:
The patent divides the processing system into multiple independent task processors that can operate in parallel on different data blocks. Each processor handles specific tasks independently, allowing simultaneous processing of multiple video frames or data blocks without sequential dependencies, thus improving overall processing capability while managing latency through parallel execution.
Solution Approach 2:
The system performs preliminary actions by pre-fetching and buffering data blocks before they are needed for processing. Data is prepared and staged in advance in memory buffers, so when a task processor needs to process a particular data block, the data is already available locally, eliminating transportation latency and allowing processors to immediately begin computation.
2Manufacturing precision
If task substitution is used to increase output quality, then quality is improved, but real-time constraint may not be met due to increased latency
Solution Approach 1:
The system dynamically selects between different task processors based on current conditions such as data characteristics, processor availability, and timing constraints. The selection is not static but adapts in real-time, choosing higher-quality processors when time permits and lower-latency processors when real-time constraints are at risk, thus balancing quality and reliability dynamically.
Solution Approach 2:
The system incorporates feedback mechanisms that monitor processing latency, quality metrics, and real-time constraint compliance. This feedback information is used to adjust task processor selection and configuration, allowing the system to learn from past performance and optimize the balance between output quality and real-time constraint satisfaction based on actual operating conditions.
Data Source
AI summary
A method including processing a stream of data in a sequence of tasks. The processing including receiving a first block of data of the stream of data, determining features associated with the first block of data, selecting, based on the features, one of a first a task to process the first block of data or a second task to process the first block of data and if the second task is selected, shift an output of the second task in time to align the output of the second task with a predicted output of the first task processing a second block of data of the stream of data.