Intelligent Cell Swapping for Frequency and Leakage Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cell designs in computer circuits face challenges in achieving a balance between higher frequency performance and lower power dissipation, as existing tools inadequately manage cell leakage, leading to poor performance and high power dissipation, and often result in mismanaged or wasted leakage.
Innovation Solution
A computer-implemented method for intelligently swapping circuit cells, which includes sorting cell classes by leakage, swapping cells to the highest or lowest leakage classes based on ceiling and floor frequency determination, and calculating effective swap weights to optimize timing paths without unnecessary high leakage cell usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If higher leakage cells are used to improve frequency performance, then frequency performance is improved, but power dissipation increases
Solution Approach 1:
The patent applies local quality by selectively assigning different leakage classes to different cells based on their specific timing requirements. Critical path cells receive higher leakage (lower Vt) cells to improve frequency, while non-critical cells use lower leakage cells to reduce power dissipation. This is achieved through the cell swapping process that replaces cells with appropriate leakage characteristics based on timing analysis.
Solution Approach 2:
The patent changes the parameter of threshold voltage (Vt) across different cells to optimize the balance between frequency and power. By swapping cells with different Vt characteristics (represented by leakage classes) into different locations in the circuit, the system achieves frequency improvement where needed while minimizing overall power dissipation.
2Device complexity
If conventional tools are used for cell selection, then design simplicity is maintained, but leakage management becomes inadequate leading to poor performance and high power dissipation
Solution Approach 1:
The patent performs preliminary action by pre-calculating timing paths and identifying critical cells before the cell swapping process. The system determines which cells need higher leakage characteristics to meet frequency targets and prepares a swapping plan that optimizes both performance and power before finalization.
Solution Approach 2:
The patent implements feedback by using timing analysis results to guide the cell swapping decisions. The system continuously evaluates timing paths, identifies cells that need swapping to meet frequency targets, and adjusts the cell assignments based on this feedback loop, thereby optimizing both frequency performance and power dissipation.
3Speed
If all cells are swapped to highest leakage class to achieve ceiling frequency, then maximum frequency is achieved, but unnecessary high leakage cell usage increases power dissipation
Solution Approach 1:
The patent applies partial action by swapping cells to highest leakage class only for the minimum necessary subset of cells required to meet the frequency target. Instead of swapping all cells, the system identifies and swaps only those cells that are critical to achieving the ceiling frequency, leaving other cells in lower leakage classes to minimize overall power dissipation.
Solution Approach 2:
The patent segments the cell population into different groups based on their timing criticality. Critical path cells are identified and swapped to highest leakage class, while non-critical cells remain in lower leakage classes. This segmentation allows the system to achieve maximum frequency where needed without incurring unnecessary power penalties across the entire circuit.
Data Source
AI summary
Embodiments of the inventive concept include a computer-implemented method for intelligently swapping circuit cells and an associated intelligent cell swapper logic section. The technique can include receiving, by an intelligent cell swapper logic section, a synthesized gate level netlist including cells each having an initial cell class. A cell class sorter can sort cell classes in order of leakage. A ceiling finder can swap the initial cell class for each of the cells to a highest cell leakage class, and determine a ceiling frequency. A floor finder can swap the highest cell leakage class for each of the cells to a lowest cell leakage class, and determine a floor frequency. An effective swap weight calculator section can determine an effective swap weight for a subset of cells based on cell attribute weighting criteria. The timing paths can be optimized to meet the ceiling frequency without unnecessarily using high leakage cells.