With the operating voltage decreasing to subthreshold, cell delay distribution tends to be a flatten, nonGaussian distribution. which makes timing analysis and optimization become more important for the integrated cir...
详细信息
With the operating voltage decreasing to subthreshold, cell delay distribution tends to be a flatten, nonGaussian distribution. which makes timing analysis and optimization become more important for the integrated cir...
详细信息
ISBN:
(数字)9798350361834
ISBN:
(纸本)9798350361841
With the operating voltage decreasing to subthreshold, cell delay distribution tends to be a flatten, nonGaussian distribution. which makes timing analysis and optimization become more important for the integrated circuit (IC) design. Although SPICE-based Monte Carlo (MC) analysis is very accurate, it is timing-consuming and thus impractical for the complex subthreshold IC design. Fast and accurate cell delay modeling is more significant. This paper proposes a parameterized cell delay modeling to predict the statistical mean(μ) and variance(σ) of subthreshold cell delay by Artificial Neural Network (ANN) model. The parameter includes both geometric parameters (i.e., transistor width and length) and operating conditions (i.e., operating voltage, temperature, input signal slew and output load capacitance). Experimental results demonstrate that the root mean square errors of μ and σ at 0.3V±10% is 3.31% and 4.34% respectively with 1100× speedup, compared with SPICE-based MC analysis. A parameterized cell delay model can predict cell delay quickly and accurately based on geometries, which is not mentioned in the prior works.
Thermal-induced warpage is a bottleneck problem in advanced packaging technology. In this paper, a new way of predicting the warpage deformation is proposed for multi-chiplet heterogeneous integration system, where th...
详细信息
Thermal issue has significant effect on the reliability and performance of integrated circuits (ICs) with the increase of integrated density, especially for 2.5-D and 3-D heterogeneous integration packaging systems. I...
详细信息
Accurate prediction for node CPU load is crucial for resource allocation in cluster. In this paper, we proposed a novel deep learning model named R-TPA-LSTM for the cluster node CPU load prediction. The proposed model...
详细信息
ISBN:
(纸本)9781665445993
Accurate prediction for node CPU load is crucial for resource allocation in cluster. In this paper, we proposed a novel deep learning model named R-TPA-LSTM for the cluster node CPU load prediction. The proposed model is composed of two components, non-linear and linear component. The non-linear component contains residual LSTM-Conv module and attention module. Residual LSTM-Conv module includes two LSTM layers with residual connection and convolutional neural network for the sake of choosing the most informative timestep in the historical window while attention module captures the relationship among different features. The goal of the linear component, which is an AR module, is to catch the drastic changes in the data. The experimental results on a real-world dataset, show that the proposed model achieves better prediction performance for CPU load than conventional models.
This paper presents a new design of a two-step Time-to-Digital Converter (TDC), which reduces the complexity of the circuits, and the power consumption and area of the circuit. The on-line self-calibration method for ...
详细信息
暂无评论