Hybrid evolutionary partitioning algorithm for heat transfer enhancement in VLSI circuits

In this paper, a new approach to improving the heat transfer in integrated circuits (ICs) is presented. It is based on improving the thermal conductivity of ICs by increasing the number of their external connections up to the level determined by the packaging standard. In order to attain this goal, a new hybrid evolutionary partitioning algorithm (HEPA) for circuits partitioning is introduced. The computations carried out for the chosen benchmarks show that HEPA is able to reach optimal solutions in the case of bipartitioning problem, and almost optimal in the case of k-way partitioning (k>2). The presented approach is especially dedicated for a flip chip interconnect technology which is used in contemporary ICs.     

An efficient macro-cell placement algorithm

A new approximation algorithm is presented for the efficient handling of large macro-cell placement problems. The algorithm combines simulated annealing with new features based on a hierarchical approach and a divide-and-conquer technique. Numerical results show that these features can lead to a considerable increase in efficiency of the placement algorithm without loss of effectiveness.     

遗传算法应用于VLSI布局的研究

本文是应用遗传算法实现ＶＬＳ门阵列模式布局,从建立相应的数学模型入手,提出一个较完备的综合布局目标函数,引入通道拥挤度的概念,从而改变了传统的常以总线长度或刻线数目的单一目标函数,使布局的构形更趋合理。在遗传３参数选取方面,提出了几种改善收性的措施,大大加快了遗传算法的收敛速度。     

Component placement in VLSI circuits using a constant pressure Monte Carlo method

When laying out a VLSI circuit on a silicon wafer the object is to pack the components of the circuit onto a wafer of minimum area subject to a variety of conflicting constraints associated with electrical interconnections among the components and input/ output connections. A constant pressure Monte Carlo method is applied to an idealized component placement problem where the object is to pack different rectangular components onto a square of minimum area with a subsidiary objective of minimizing the total length of wires interconnecting the components. The Monte Carlo method is found to be remarkably effective in solving this idealized problem. No other method for solving this problem is known.     

An improved algorithm for accelerating reconfiguration of VLSI array

Reducing the number of visits to failure-free nodes can effectively reduce the reconstruction time of logical columns and improve the reconstruction efficiency. In this paper, we describe a new method to speed up the reconfiguration for the VLSI arrays. An efficient algorithm was proposed based on shortest path first principle for accelerating reconfiguration of VLSI processor subarrays with high power efficiency to meet the requirement of the power consumption of embedded system. The proposed algorithm greatly reduces the number of visits to the fault-free PEs for constructing a local optimal logical column and effectively reduces the construction time. Experimental results show that the proposed algorithm is capable of reducing the consumption time by 32.15% and reducing the numbers of visited PEs by 49.61% for a 128 × 128 host array with 20% falut rate.     

An algorithm ofk-line location for BBL in LSI/VLSI

In this paper, aK-line location algorithm for building block cells in LSI/VLSI is presented. When the relative positions of rectangular cells are given, there are 2 states according to the two orientations of a cell. It is proved that to find the optimum solution from the 2^N states can be reduced to calculate theN states inK-line algorithm. So the algorithm is shown very effective and can be used with association for cluster method in BBL placement. Under certain conditions, this method can also be used to pesudo BBL placement directly.     

基于改进遗传算法的多目标JSP调度研究

针对多目标车间作业调度问题(JSP),提出了一种混合遗传算法,将多目标遗传算法得出的初步优化结果作为粒子群算法的初始粒子,利用粒子群算法强化局部搜索,加快收敛速度,改善了简单遗传算法局部搜索能力差、迭代效率低的问题.仿真结果表明了该算法对JSP调度的良好效果.     

基于信息熵的改进PESA算法

针对PESA算法所需的计算运算量、计算难度及运算时间都随着解集数量的增加而急剧增加的问题,将熵值度量指标引入到PESA算法中,提出了基于信息熵的PESA算法(C-PESA, comentropy-based PESA)。该算法根据信息熵指标在量化度量Pareto解集的分布特性,判断种群进化是否到达成熟阶段,本算法迭代1 300次时即到达成熟阶段,从而尽早结束了算法复杂的优化过程,在一定程度上简化了PESA算法的时间复杂度。仿真结果表明,随着进化种群数量的增长,C-PESA算法的计算量只是呈现线性增加,算法的计算时间缩短接近4倍,进化计算效率得到提高。     

一种改进变异算子的混合遗传算法

分析了遗传算法传统变异算子的缺陷,为解决遗传算法搜索效率低下及早熟收敛的问题,设计了一种融合遗传搜索和模式搜索的混合遗传算法。理论分析与实验仿真结果表明,所给出的混合遗传算法是有效的,在收敛速度、精度和稳定性方面均有明显的提高。     

基于遗传算法的VLSI布图规划方法

提出了一种有效的基于遗传算法的VLSI布图规划方法.在染色体的表达中,对软模块不同形状和硬模块的布局方向进行了编码,并设计了有效的启发式解码方法进行解码.测试结果表明,该算法比已有算法得到了更优的结果.     

基于均匀设计的多目标自适应遗传算法及应用

提出一种多目标遗传算法,将均匀设计技术应用于适应度函数合成和交叉算子构造,以提高遗传算法的空间搜索均匀性、子代质量和运算效率.分析和实验结果表明,该方法可缩短算法运行时间和得到分布较均匀的Pareto有效解集;配合基于元件标称值的网表级高效编码方案和考虑基因位差异的遗传概率调整策略,可实现模拟电路自动设计,通过单次运行即获得对应不同偏好的多种实用化设计结果.     

Multidisciplinary heat generating logic block placement optimization using genetic algorithm

A multidisciplinary optimization methodology for placement of heat generating semiconductor logic blocks on integrated circuit chips is presented. The methodology includes thermal and wiring length criteria, which are optimized simultaneously using a genetic algorithm. An effective thermal performance prediction methodology based on a superposition method is used to determine the temperature distribution on a silicon chip due to multiple heat generating logic blocks. Using the superposition method, the predicted temperature distribution in the silicon chip is obtained in much shorter time than with a detailed finite element model and with comparable accuracy. The main advantage of the present multidisciplinary design and optimization methodology is its ability to handle multiple design objectives simultaneously for optimized placement of heat generating logic blocks. Capabilities of the present methodology are demonstrated by applying it to several standard benchmarks. The multidisciplinary logic block placement optimization results indicate that the maximum temperature on a silicon chip can be reduced by up to 7.5 °C, compared to the case in which only the wiring length is minimized.     

一种时延驱动的VLSI布局方法

时廷特性对于高性能的超大规模集成电路(VLSI)来讲是十分重要的。本文提出了一个新的时延驱动的布局方法。在初始布局中,我们提出了给线网加权的新方法,在迭代改善布局中提出了等位场的概念。实验结果表明:这是一种有效的时延驱动布局方法。     

An evolutive framework for server placement optimization to digital twin networks

Digital twin network (DTN) is a foremost enabler for efficient optimization in modern networks, as it owns massive real-time data and requires interaction with the physical network in real-time. When constructing a DTN, it is necessary to deploy many servers in the physical network for digital models' storage, calculation, and communication. Evolutionary algorithms show outstanding global optimization capabilities compared to the constructive heuristic method in such an optimization problem. However, due to the high dimensionality of the problem and the complicated evaluation of the deployment plan, evolutionary algorithms easily fall into the optimum local at a high computational cost, given that the server placement problem is an NP-hard combinatorial optimization problem. In this research, we propose an evolutionary framework for server layout optimization that significantly improves the optimization efficiency of evolutionary algorithms and reduces the algorithm's computational cost. An offline-learning-based approach is used to reduce the search space, and a self-examining guided local search method is proposed to improve the search efficiency. Additionally, a look-up table-based hybrid approach is used for solution evaluation, reducing computational overhead. Experimental results show that the proposed framework and optimization strategy can significantly improve the evolutionary algorithm search efficiency and achieve excellent convergence performance.     

基于混合粒子群算法求解TSP问题

遗传算法是研究TSP问题中最为广泛的一种算法,它具有全局搜索的能力。而粒子群算法收敛速度较快,但容易造成局部最优的情况。本文基于遗传算法的交叉变异设计了混合粒子群算法,通过对TSP问题求解分析,证实该方法提高了标准粒子群的搜索能力,获得了较高的收敛速度和近似最优解。     

Shamir-based virtual machine placement policy

In order to alleviate the harm of co-location attack in cloud environment,a virtual machine placement strategy based on Shamir was proposed,and a virtual machine placement framework was designed.The security of key data was ensured by the blockchain in the proposed placement strategy.The virtual machine placement policy could effectively improve the security of the virtual machine and the load balancing capability of the cloud environment,and reduce resource waste.Finally,the effectiveness of the scheme is proved by simulation experiments.     

用模拟退火算法实现集成电路热布局优化

介绍了一种综合考虑集成电路电学性能指标以及热效应影响的布局优化方法 .在保证传统设计目标 (如芯片面积、连线长度、延迟等 )不被恶化的基础上 ,通过降低或消除芯片上的热点来优化集成电路芯片的温度分布情况 ,进而优化整个电路性能 .并将改进的模拟退火算法应用于集成电路的热布局优化 ,模拟结果表明该方法与传统布局方法相比在保持了较好的延迟与连线长度等设计目标的同时 ,很好地改善了芯片表面的热分配情况     

基于正交设计的元胞多目标遗传算法

元胞多目标遗传算法在求解两目标优化问题时是比较高效的.但是,初步实验显示其在求解三目标优化问题(例如DTLZ系列)时,表现不是十分令人满意.为了进一步提高算法的性能,引入了正交设计的思想,提出了基于正交设计的多目标元胞遗传算法.在改进算法的迭代过程中,先对父代个体进行分段,之后按照正交表来对这些片段进行重新组合产生多个子代个体,然后从这些子代个体中找出适应度较优的进入下一代种群.实验结果表明,引入正交设计思想能够提高算法性能,与其他优秀算法进行比较的结果说明,改进算法求解三目标问题(DTLZ系列)也是具有竞争力的.     

并行算法在VLSICAD中的应用

本文简介并算法在ＶＬＳＩＣＡＤ中的应用，介绍了这一领域的研究现状，所采用的关键技术以及进一步的研究发展方向。     

基于遗传算法的声表面波滤波器进化设计

传统方法设计声表面波滤波器的主要缺点是设计过程复杂.本文提出了一种新的设计方法,它将声表面波滤波器的所有指条结构特征的集合作为染色体,并分成两个基因段分别编码,采用自适应调整策略调整进化过程中两基因段的交叉概率和变异概率,通过遗传操作自动设计出综合性能指标最优的声表面波滤波器.相对于同指对数未加权声表面波滤波器,中心频率插入损耗损失不多(-2dB),而旁瓣抑制约改善了26dB.