Multiprocessor Memory Allocation for Activity Sequences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multiprocessor systems, existing methods for implementing UML activity diagrams do not efficiently manage memory allocation and block assignment during runtime, leading to potential interruptions or slowdowns when activities require memory buffers, as these processes are typically performed sequentially and can interfere with data processing.
Innovation Solution
A multiprocessor computer architecture that interweaves the performance of activity sequences with block assignment processes, allowing for just-in-time allocation and release of memory blocks, ensuring that memory allocation is completed before activity sequences commence, and unused blocks are identified and powered down to conserve resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If memory allocation and block assignment are performed sequentially before activity sequences, then memory resources are properly allocated, but processing time is increased and performance is hindered
Solution Approach 1:
The patent applies preliminary action by performing memory allocation and block assignment just-in-time before each activity sequence commences, rather than allocating all memory upfront or during runtime interruptions. This ensures memory is ready when needed without hindering overall processing speed.
Solution Approach 2:
The patent enables continuous processing by interweaving memory allocation with activity execution seamlessly. Memory allocation occurs in the background during transitions between activities rather than interrupting the data processing flow, maintaining continuous useful action.
2Reliability
If memory blocks are allocated in advance for all activities, then sufficient memory resources are available, but unused memory blocks consume energy unnecessarily
Solution Approach 1:
The patent applies local quality by allocating memory blocks specifically to activities based on their individual requirements rather than providing universal memory allocation to all activities. Each activity receives memory blocks locally assigned just-in-time, ensuring no unused blocks consume energy.
Solution Approach 2:
The patent implements discarding and recovering by releasing memory blocks back to the pool after each activity completes its execution. This allows memory to be dynamically reused for subsequent activities rather than being permanently allocated, reducing energy consumption from unused blocks.
3Adaptability or versatility
If block assignment is performed during activity execution, then memory can be allocated dynamically, but processing is interrupted or slowed down
Solution Approach 1:
The patent applies preliminary action by completing all block assignment operations before an activity sequence commences. This preliminary preparation ensures dynamic memory allocation is achieved without interrupting the actual processing execution.
Solution Approach 2:
The patent introduces an intermediary layer between memory management and activity execution. The memory management system operates in the background during transitions, acting as an intermediary that prepares memory without directly interfering with the speed of data processing activities.
Data Source
AI summary
A method and apparatus for processing data by a computer and a method of determining data storage requirements of a computer for earning out a data processing task.