Optimization of Individual Well Adaptive Body Biasing (IWABB) Using a Multiple Objective Evolutionary Algorithm

Gregg, Jessica; Chen, T.W.

doi:10.1109/isqed.2005.87

Cited by 10 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our placement-aware work relies on the optimized clustering of gates to reduce the number of required on-die body biases to a small number (2)(3)(4). In comparison to the traditional technique of dual Vth assignment, we show that our physical design aware ABB approach can produce designs with 2-9X tighter delay distributions and power reductions of 38-71% while tightly controlling area, wirelength and bias routing overheads.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

A statistical framework for post-silicon tuning through body bias clustering

Kulkarni

Sylvester

Blaauw

2006

Proceedings of the 2006 IEEE/ACM International Conference on Computer-Aided Design - ICCAD '06

View full text Add to dashboard Cite

Adaptive body biasing (ABB) is a powerful technique that allows post-silicon tuning of individual manufactured dies such that each die optimally meets the delay and power constraints. Assigning individual bias control to each gate leads to severe overhead, rendering the method impractical. However, assigning a single bias control to all gates in the circuit prevents the method from compensating for intra-die variation and greatly reduces its effectiveness. In this paper, we propose a new variability-aware method that clusters gates at design time into a handful of carefully chosen independent body bias groups, which are then individually tuned post-silicon for each die. We show that this allows us to obtain near-optimal performance and power characteristics with minimal overhead. For each gate, we generate the probability distribution of its post-silicon ideal body bias voltage using an efficient sampling method. We then use these distributions and their correlations to drive a statistically-aware clustering technique. We study the physical design constraints and show how the area and wirelength overhead can be significantly limited using the proposed method. Compared to a fixed design time based dual threshold voltage assignment method, we improve leakage power by 38-71% while simultaneously reducing the standard deviation of delay by 2-9X.

show abstract

Section: Discussionmentioning

confidence: 99%

“…However, the body voltages are fixed at design time using a deterministic formulation and post-silicon tuning is not considered. In [4], results for two small ABB enabled designs are presented. This work relies on a multiple objective evolutionary algorithm to determine ABB voltages for individual device wells post-silicon.…”

Section: Introductionmentioning

confidence: 99%

A statistical framework for post-silicon tuning through body bias clustering

Kulkarni

Sylvester

Blaauw

2006

Proceedings of the 2006 IEEE/ACM International Conference on Computer-Aided Design - ICCAD '06

View full text Add to dashboard Cite

show abstract

“…Very‐large‐scale integration design has also benefited from EAs. In Gregg and Chen solved the problem of adjusting the body bias voltage. Because of the reduction in the technology, the process variations problems are growing.…”

Section: Electronic Design Automation Toolsmentioning

confidence: 99%

A review of bioinspired computer‐aided design tools for hardware design

Lanchares¹,

Garnica²,

Fernández-de-Vega

et al. 2013

Concurrency and Computation

View full text Add to dashboard Cite

During the last two decades, bioinspired techniques have emerged as a powerful optimization and problem solving tool for a wide range of real world applications. Parallel architectures and computer-aided design tools have been also influenced by this evolutionary fashion. In this paper, we give a broader view of the application of techniques inspired by nature to hardware design and parallel architectures problem solving. Our aim is to furnish an overview of the various bioinspired techniques based tools that have been used so far to solve the problems of automatic hardware design. We can claim that a lot of the approaches found in the literature suffer from a lack of interdisciplinary interaction among researchers of both evolutionary computation and hardware design fields. In addition, we have also detected that some multi-objective problems do not use the appropriate algorithms.Among widely known heuristics, gradient algorithms, such as Hill Climbing, were soon employed. These algorithms start the search process from a single point, stochastically generated from the solution space, and then perform a search process using the direction of the gradient towards the objective function. These algorithms are of an easy implementation and relatively efficient. However, local optima are usually the output and no general procedures exist to know the distance from the global optimum.An improvement over the previous approach was obtained by the simulated annealing (SA) algorithm [2], capable of removing some of the previously described disadvantages. SA begins with a randomly generated candidate solution-a point in the search space-with some specific features, and then, the algorithm proceeds by applying a disturbance to the candidate solution, thus trying to improve it. If the solution improves, it is selected and becomes the next candidate solution; otherwise, a probability decides whether to continue with the initial candidate or with the new one. The probability of accepting a worse solution depends on the system temperature-a fundamental parameter for SA. The system starts with a very high temperature, so most of the worse solutions are accepted at the beginning. The temperature is then lowered gradually; the lower the temperature, the lower the probability of accepting a worse solution. So far, SA has been the most frequently employed heuristic in computer-aided design (CAD) tools.Nevertheless, a new kind of heuristic has gained popularity: bioinspired algorithms (BAs) [3]. Among BAs, evolutionary algorithms (EAs) try to emulate the natural evolution of species, following Darwin's principles of natural selection. One of their main advantages is the employment of a set (population) of candidate solutions (individuals) at the same time, which gives rise to an intrinsically parallel search process. We can also find other bioinspired approaches on the basis of the social behavior of the species such as particle swarm optimization (PSO) [4] or ant colony optimization (ACO) [5] algorithms.This review focuses on the appli...

show abstract

“…As opposed to [8], we allow different PEs in the system to operate at different clock speeds, based on the body-bias voltage of the island they lie in, but constrain the execution deadlines of the task graph running on the multiprocessor system. In [6], the authors also consider the case of systems partitioned into a number of body-bias clusters; however, the focus of this work is on the allocation of body-bias voltage assignments for each die after they have been manufactured. It does not address the design time partitioning of the system into body-bias islands.…”

Section: Related Workmentioning

confidence: 99%

“…It does not address the design time partitioning of the system into body-bias islands. Furthermore, like [8], [6] concentrates on fully synchronous designs with constraints on the global clock frequency of the system, while we impose constraints only on system-level performance metrics.…”

Section: Related Workmentioning

confidence: 99%

System-level mitigation of WID leakage power variability using body-bias islands

Garg

Marculescu

2008

Proceedings of the 6th IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis

View full text Add to dashboard Cite

Adaptive Body Biasing (ABB) is a popularly used technique to mitigate the increasing impact of manufacturing process variations on leakage power dissipation. The efficacy of the ABB technique can be improved by partitioning a design into a number of "body-bias islands," each with its individual body-bias voltage. In this paper, we propose a system-level leakage variability mitigation framework to partition a multiprocessor system into body-bias islands at the processing element (PE) granularity at design time, and to optimally assign body-bias voltages to each island post-fabrication. As opposed to prior gate-and circuit-level partitioning techniques that constrain the global clock frequency of the system, we allow each island to run at a different speed and constrain only the relevant system performance metrics -in our case the execution deadlines. Experimental results show the efficacy of the proposed framework in reducing the mean and standard deviation of leakage power dissipation compared to a baseline system without ABB. At the same time, the proposed techniques provide significant runtime improvements over a previously proposed Monte-Carlo based technique while providing similar reductions in leakage power dissipation.

show abstract

Optimization of Individual Well Adaptive Body Biasing (IWABB) Using a Multiple Objective Evolutionary Algorithm

Cited by 10 publications

References 6 publications

A statistical framework for post-silicon tuning through body bias clustering

A statistical framework for post-silicon tuning through body bias clustering

A review of bioinspired computer‐aided design tools for hardware design

System-level mitigation of WID leakage power variability using body-bias islands

Contact Info

Product

Resources

About