Cell Placement in Multi-Level Routing Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current electronic design automation (EDA) systems face challenges in efficiently generating and updating the placement of cells in multi-level routing trees, particularly in satisfying design rule constraints while minimizing physical resources like wire capacitance, which can be computationally expensive due to the use of undirected algorithms.
Innovation Solution
The implementation of routing algorithms to identify connections between node placement points, allowing for the evaluation of a limited number of cell locations that meet design rule constraints, and the use of routing regions to optimize cell placement by reducing the number of possible placement points, thereby minimizing wire length and power usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If undirected algorithms are used to test node placement, then all design rule constraints can be satisfied, but computing resources and computation time become prohibitively expensive
Solution Approach 1:
The patent segments the routing tree into directed levels (parent nodes, child nodes, grandchild nodes) and processes placement constraints directionally rather than exhaustively. By dividing the search space into hierarchical levels with specific constraint applications at each level, the algorithm achieves constraint satisfaction without requiring computationally expensive undirected search of all possible placements.
Solution Approach 2:
The patent applies preliminary actions by pre-establishing routing subregions and connecting routes before final placement optimization. Routing subregions are generated to define valid placement areas, and connecting routes are pre-computed to establish wire length constraints. This preliminary structuring reduces the search space for final placement, avoiding the need for exhaustive undirected algorithms while ensuring all constraints are met.
2Loss of energy
If routing algorithms with routing subregions and connecting routes are used, then wire length and power usage are minimized, but the complexity of the routing structure increases
Solution Approach 1:
The patent introduces routing subregions and connecting routes as intermediary structures that mediate between node placement points and final routing decisions. These intermediaries break down the complex routing problem into manageable components: routing subregions define valid placement areas, connecting routes establish optimized wire paths, and together they enable systematic optimization of wire length and power usage without requiring overly complex routing algorithms.
Solution Approach 2:
The patent optimizes wire length and power usage by changing key routing parameters such as wire capacitance values, routing region boundaries, and connecting route paths. By adjusting these parameters based on timing analysis and placement constraints, the system achieves energy efficiency while maintaining manageable routing structure complexity through parameter optimization rather than structural complexity.
3Productivity
If a limited number of cell locations are evaluated based on routing regions, then computation costs are reduced, but the thoroughness of placement optimization may be compromised
Solution Approach 1:
The patent applies local quality by evaluating placement options with different levels of thoroughness in different regions. Routing subregions are identified as areas with specific placement constraints, and within each subregion, a limited but strategically selected set of cell locations is evaluated. This localized approach ensures that placement optimization is thorough enough for each specific region while maintaining overall computation efficiency, as not all possible locations need to be evaluated uniformly.
Data Source
AI summary
Systems, methods, media, and other such embodiments are described for placement of cells in a multi-level routing tree, where placement of a mid-level parent node between a grandparent node and a set of child nodes is not set. One embodiment involves generating a first routing subregion between a first set of child nodes associated with a first grandparent node and a first connecting route from the first routing subregion to the first grandparent node, which together are set as a first routing region comprising the first routing subregion and the first connecting route. Sampling points are selected along the first routing region, and for each sampling point a set of operating values associated with the sampling point is calculated. A position for the parent node is selected based on the operating values for the sampling points.