SOFTWARE POWER ANALYSIS AND OPTIMIZATION FOR POWER-AWAREMULTICORE SYSTEMSbySHINANWANGMay 2014 Advisor: Dr. Weisong ShiMajor: Computer ScienceDegree: Doctor of Philosophy Among all the factors in sustainable computing, power dissipation and energy consump-tion, arguably speaking, are fundamental aspects of modern computer systems. Different fromperformance metric,...

| Decreasing capacitance of bus lines is one of the e ective ways to reduce whole power dissipation of LSIs. In this paper we compare microprocessors designed based on a bus architecture and a multiplexer architecture in terms of power dissipation and delay time. Through implementation of a test chip, the multiplexer architecture is e ective to reduce power dissipation by about 30%.

Motivation for reducing power dissipation during test application is presented. A scheme for reducing power dissipation during test application, when scan test structure is used, is proposed. Algorithms required to exploit the proposed technique are discussed. Experimental results are presented. keywords: Power dissipation, Full Isolated Scan, Full Integrated Scan.

P ower dissipation has evolved into an instrumental design optimization objective due to the growing demand for portable electronics equipment as well as due to excessive heat generation in high-performance systems. In the former case , low-power techniques are employed to prolong battery life, while in the latter case, low-power techniques are required to mitigate the reliability problems that...

This work provides a design for two types of sensors, based on the thermal dissipation and heat ratio methods of sap flow calculation, for moderate to large scale deployments for the purpose of monitoring tree transpiration. These designs include a procedure for making these sensors, a quality control method for the final products, and a complete list of components with vendors and pricing info...

In this paper, we discuss a design technique to reduce the power dissipation of CML buffers for on-chip transmissionlines. CML buffers can operate in higher frequency than conventional static CMOS buffers. On the other hand, the power dissipation is larger than that of the static CMOS buffers. We reduce the power dissipation by using an impedance-unmatched driver. From the pole frequency analys...

Two exact results are presented for the thermal dissipation in a Rayleigh-BCnard geometry. The first result expresses the conditional dissipation at each value of 0 within the entire volume of the system in terms of the thermal flux and the PDF for 19. The second gives the conditional dissipation in a region of the cell in terms of the conditional flux into the region. These results are used to...

This paper concerns the one dimensional problem of the porous-thermo-elasticity. Two kinds of dissipation process are considered: the viscosity type in the porous structure and the thermal dissipation. It is known that when only thermal damping is considered or when only porous damping is considered we have the slow decay of the solutions. Here we prove that when both kinds of dissipation terms...

In the ultra low power end of design spectrum when performance is of secondary importance, digital subthreshold logic circuits are more applicable than the regular MOS logic. In this paper, we propose two different subthreshold logic families: 1) variable threshold voltage subthreshold CMOS (VT-Sub-CMOS) and 2) subthreshold dynamic threshold voltage MOS (SubDTMOS) logic. Both these logic famili...

Context. Visco-resistive damping in line-tied magnetic X-points is examined. Aims. The goal is to determine whether fast, Alfvénic energy dissipation is possible for X-point disturbances damped by the plasma resistivity and non-isotropic viscosity. Methods. The response of X-points to planar and axial perturbations is explored numerically by solving the linearized compressible MHD equations in ...