China
Africa
Explore chapters and articles related to this topic
Digital Systems
Published in Wai-Kai Chen, Analog and VLSI Circuits, 2018
Festus Gail Gray, Wayne D. Grover, Josephine C. Chang, Bing J. Sheu, Roland Priemer, Rung Yao, Flavio Lorenzelli
Advances in very large-scale integrated (VLSI) processing technology, particularly CMOS, have resulted in nanometer-scale processes with applications at clock speeds of several gigahertz. For example, at the time of this revision the state of the art is fairly well represented by the Intel Core2 Duo processor chip, which is implemented in 65 nm CMOS, consists of 376 million transistors, and clocks at 2.66 GHz. New design challenges must be mastered to realize systems at an ever-increasing clocking rates and circuit sizes. In particular, clocking-related issues of skew, delay, power dissipation, and switching noise can be design-limiting factors. In large synchronous designs, the clock net is typically the largest contributor to on-chip power dissipation and electrical noise generation, particularly “ground bounce,” which reduces noise margin. Ground bounce is a rise in ground potential due to surges of current returning through a nonzero (typically inductive) ground path impedance. At the board and shelf level, clock distribution networks can also be a source of electromagnetic emissions, and may require considerable delay tuning for optimization of the clock distribution network.
Minimum-lap-time optimisation and simulation
Published in Vehicle System Dynamics, 2021
The combination of a QSS paradigm and a transient vehicle model is described in [43], in which the trajectory is pre-defined. During the calculation of the optimal MLTS strategy, the transient states are treated as distance-dependent parameters. The QSS outcome is used post-facto to calculate a revised dynamic response, which induces, in turn, a revised QSS solution. This iterative process is repeated until the dynamic states have settled. The technique accommodates dynamics effects such as the tyres' damping and temperature – a grip scaling factor is used that depends on the tyre surface temperature, and a cornering stiffness factor that depends on the tyre core temperature. To be clear, the method does not use pre-computed g-g diagrams, instead, the states of the QSS model are computed step-by-step when solving an NLP problem. The given application considers a 2009 Le Mans prototype on the Circuit de la Sarthe. The computing time is approximately 12 minutes on a 2.5 GHz Intel core 2 workstation.
Methodology for Voltage Stability Analysis Using Hopf Bifurcation and Continuation Power Flow Simulator
Published in Electric Power Components and Systems, 2020
Matheus M. Roque, José Eduardo O. Pessanha
To gain insight into the computational cost of other methods, let us focus on the direct method proposed in [21], which is fast even compared to other direct methods. The authors developed their proposal in Matlab and investigated the same power systems as here, but with slightly different data. The simulations were run in a computer Intel Core 2 1.86 GHz processor and 4GB of RAM. For the 2-area 11-bus power system, their method obtained a CPU time of 0.31 ms, whereas a classical method of 8.16 ms. It will be recalled that our proposal is not intended for online studies, and therefore cannot compete with these methods in terms of CPU time. Our proposal aims at a simple implementation, and to correctly interact with a continuation power flow for finding the HBP and positive eigenvalues. Besides, the actual version has not yet been tested to compute saddle-node bifurcation (future investigations will include this type of bifurcation).
Scheduling multimedia services in cloud computing environment
Published in Enterprise Information Systems, 2018
Yunchang Liu, Chunlin Li, Youlong Luo, Yanling Shao, Jing Zhang
The simulations run on a machine with an Intel Core 2 Duo CPU and 4 GB RAM. Each incoming task is set to have 200/400 MB input data size, 106–107 Million Instructions and 0–1.0 trust requirement. The processor speed follows a uniform distribution in the interval [800, 1000] MIPS. The pricing model of resources based on Amazon prices, is summarized in Table 2. The failure rate follows a uniform distribution in the interval [0, 0.050].