Logical Dispatchable Unit Engine for Heterogeneous Processor Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in effectively dispatching units of work between general processors and specialty engines in multi-threaded computer systems, as they often require handling both specialized and general processing tasks efficiently, leading to increased hardware costs and performance throttling.
Innovation Solution
A logical dispatchable unit engine identifies units of work and determines whether to dispatch them to general processors or specialty engines, communicating with the operating system kernel to optimize execution, thereby reducing the workload on general processors and enhancing performance by utilizing specialty engines for optimized tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If specialty engines are used to execute specialized units of work, then processing efficiency is improved, but device complexity increases
Solution Approach 1:
The patent segments the processing system into general processors and specialty engines, with a dispatchable unit engine acting as an intermediary layer. This segmentation allows specialized tasks to be routed to appropriate engines while maintaining system manageability through modular architecture.
Solution Approach 2:
The dispatchable unit engine serves as an intermediary between the operating system and the heterogeneous processor architecture. It abstracts the complexity of routing decisions by providing a standardized interface for dispatching units of work to either general processors or specialty engines based on their characteristics.
2Adaptability or versatility
If more general processors are added to handle both specialized and general tasks, then adaptability is improved, but hardware costs increase
Solution Approach 1:
The dispatchable unit engine provides universal functionality by managing both specialized and general processing tasks through a unified dispatching mechanism. This allows the system to handle diverse workloads without requiring separate processing paths, reducing the need for additional hardware.
Solution Approach 2:
The patent creates virtual copies of processing capabilities through the dispatchable unit engine, which can simulate or emulate specialized processing functions on general processors when specialty engines are unavailable or inappropriate, reducing hardware requirements.
3Speed
If units of work are dispatched to specialty engines, then processing speed is improved, but loss of information increases due to exception handling complexity
Solution Approach 1:
The dispatchable unit engine implements feedback mechanisms to track the execution status of units of work across different engine types. When exceptions occur, the system uses feedback information to determine appropriate recovery actions, such as retrying on a different engine type or escalating to higher-level error handling.
Solution Approach 2:
The system prepares exception handling protocols and recovery procedures in advance, before exceptions actually occur. This includes pre-configuring retry logic, fallback processing paths, and error recovery mechanisms that are activated only when needed, minimizing information loss during exception handling.
Data Source
AI summary
A Logical Dispatchable Unit Engine identifies a unit of work to be executed. The Logical Dispatchable Unit Engine determines whether the unit of work is to be dispatched to a general processor or a specialty engine based on characteristics associated with the unit of work, instructions associated with the unit of work and/or restrictions associated with the unit of work. The Logical Dispatchable Unit Engine communicates with an Operating System kernel to dispatch the unit of work for execution, responsive to determining whether the unit of work is to be dispatched to a general processor or a specialty engine.