When designing a computer system, benchmark programs are used with cycle accurate performance/power simulators and HDL level simulators to evaluate novel architectural enhancements, perform design space exploration, understand the worst-case power characteristics of various designs and find performance bottlenecks. This research effort is directed towards automatically generating synthetic benchmarks to tackle three design challenges: 1) For most of the simulation related purposes, full runs of modern real world parallel applications like the PARSEC, SPLASH suites cannot be used as they take machine weeks of time on cycle accurate and HDL level simulators incurring a prohibitively large time cost 2) The second design challenge is that, some of these real world applications are intellectual property and cannot be shared with processor vendors for design studies 3) The most significant problem in the design stage is the complexity involved in fixing the maximum power consumption of a multicore design, called the Thermal Design Power (TDP). In an effort towards fixing this maximum power consumption of a system at the most optimal point, designers are used to hand-crafting possible code snippets called power viruses. But, this process of trying to manually write such maximum power consuming code snippets is very tedious. All of these aforementioned challenges has lead to the resurrection of synthetic benchmarks in the recent past, serving as a promising solution to all the challenges. During the design stage of a multicore system, availability of a framework to automatically generate system-level synthetic benchmarks for multicore systems will greatly simplify the design process and result in more confident design decisions. The key idea behind such an adaptable benchmark synthesis framework is to identify the key characteristics of real world parallel applications that affect the performance and power consumption of a real program and create synthetic executable programs by varying the values for these characteristics. Firstly, with such a framework, one can generate miniaturized synthetic clones for large target (current and futuristic) parallel applications enabling an architect to use them with slow low-level simulation models (e.g., RTL models in VHDL/Verilog) and helps in tailoring designs to the targeted applications. These synthetic benchmark clones can be distributed to architects and designers even if the original applications are intellectual property, when they are not publicly available. Lastly, such a framework can be used to automatically create maximum power consuming code snippets to be able to help in fixing the TDP, heat sinks, cooling system and other power related features of the system. The workload cloning framework built using the proposed synthetic benchmark generation methodology is evaluated to show its superiority over the existing cloning methodologies for single-core systems by generating miniaturized clones for CPU2006 and ImplantBench workloads with only an average error of 2.9% in performance for up to five orders of magnitude of simulation speedup. The correlation coefficient predicting the sensitivity to design changes is 0.95 and 0.98 for performance and power consumption. The proposed framework is evaluated by cloning parallel applications implemented based on p-threads and OpenMP in the PARSEC benchmark suite. The average error in predicting performance is 4.87% and that of power consumption is 2.73%. The correlation coefficient predicting the sensitivity to design changes is 0.92 for performance. The efficacy of the proposed synthetic benchmark generation framework for power virus generation is evaluation on SPARC, Alpha and x86 ISAs using full system simulators and also using real hardware. The results show that the power viruses generated for single-core systems consume 14-41% more power compared to MPrime on SPARC ISA. Similarly, the power viruses generated for multicore systems consume 45-98%, 40-89% and 41-56% more power than PARSEC workloads, running multiple copies of MPrime and multithreaded SPECjbb respectively. / text

Effective application of thermal interface materials (TIM) sandwiched between silicon and a heat spreader in a microelectronic package for improved heat dissipation is studied through thermal and mechanical characterization of high thermally conductive carbon nanotubes (CNTs) integrated into eutectic gallium indium liquid metal (LM) wetting matrix. Thermal conductivity data from Infrared microscopy tool reveals the dependence of experimental factors such as matrix types, TIM contacting interfaces, orientation of CNTs and wetting of CNTs in the matrix on the thermal behavior of TIM composite. Observed generalized trend on LM-CNT TIM shows progressive decrease in effective thermal conductivity with increasing CNT volume fractions. Further thermal characterizations LM-CNT TIM however show over 2x increase in effective thermal conductivity over conventional polymer TIMs (i.e. TIM from silicone oil matrix) but fails to meet 10x improvement expected. Poor wetting of CNT with LM matrix is hypothesized to hinder thermal improvement of LM-CNT TIM composite. Thus, wetting enhancement technique through electro-wetting and liquid crystal (LC) based matrix proposed to enhance CNT-CNT contact in LM-CNT TIM results in thermal conductivity improvement of 40 to 50% with introduction of voltage gradient of 2 to 24 volts on LM-CNT TIM sample with 0.1 to 1 percent CNT volume fractions over non voltage LM-CNT TIM test samples. Key findings through this study show that voltage tests on LC- CNT TIM can cause increased CNT-CNT networks resulting in 5x increase in thermal conductivity over non voltage LC-CNT TIM and over 2x improvement over silicone-CNT TIMs. Validation of LM wetting of CNT hypothesis further shows that wetting and interface adhesion mechanisms are not the only factors required to improve thermal performance of LM-CNT TIM. Anisotropic characteristic of thermal conductivity of randomly dispersed CNTs is a major factor causing lower thermal performance of LM-CNTs TIM composite. Other factors resulting in LM-CNT TIM decreasing thermal conductivity with increasing CNT loading are (i) Lack of CNT-CNT network due to large difference in surface tension and mass density between CNTs and LM in TIM composite (ii) Structural stability of MWCNT and small MFP of phonons in ~5um MWCNTs compared to the system resulted in phonon scattering with reduced heat flow (iii) CNT percolation threshold limit not reached owing to thermal shielding due to CNT tube interfacial thermal resistance. While mixture analytical models employed are able to predict thermal behaviors consistent with CNT-CNT network and CNT- polymer matrix contact phenomenon, these models are not equipped to predict thermo-chemical attributes of CNTs in LM-CNT TIM. Extent of LM-CNT wetting and LM-solid surface interfacial contact impacts on interfacial thermal resistance are investigated through LM contact angle, XPS/AES and SEM-EDX analyses on Au/Ni and Ni coated copper surfaces. Contact angle measurements in the range of 120o at both 55oC and 125oC show non wetting of LM on CNT, Au and Ni surfaces. Interface reactive wetting elemental composition of 21 days aged LM on Au/Ni and Ni surfaces reveals Ga dissolution in Au and Ni diffusion of ~0.32um in Au which are not present for similar analysis of 1 day LM on Au/Ni surface. Formation of Au-Ni-Ga IMC and IMC-oxide iono-covalency occurrence at the interface causes reduction in surface tension and reduction in interfacial contact resistance.

The thesis aims to propose a system for heating and hot water for a family house, according to drawings supplied by the building architect. At the request of the investor as the primary object will be considered a variant of underfloor heating. For comparing the heat loss, economic and possibly to State subsidies system selected will be processed by an alternative option to the ventilation air heating. Underfloor heating and hot air are among the low-temperature heating system, which can use low temperature heat sources, such as heat pumps. This assumption corresponds to investor requests, the source of heat in a family house with a heat pump and gas boiler. For hot water will be used as an alternative to a fireplace in the living room. To use the results of the thesis is to calculate the heat loss and heat, according to current standards include design and hot-water heating system including hot-drawing documentation. Finally, the economic evaluation of different alternatives in terms of investment and operating costs.