Prediction-Based Power-Performance Adaptation of Multithreaded Scientific Codes

Computing has recently reached an inflection point with the introduction of multicore processors. On-chip thread-level parallelism is doubling approximately every other year. Concurrency lends itself naturally to allowing a program to trade performance for power savings by regulating the number of active cores; however, in several domains, users are unwilling to sacrifice performance to save power. We present a prediction model for identifying energy-efficient operating points of concurrency in well-tuned multithreaded scientific applications and a runtime system that uses live program analysis to optimize applications dynamically. We describe a dynamic phase-aware performance prediction model that combines multivariate regression techniques with runtime analysis of data collected from hardware event counters to locate optimal operating points of concurrency. Using our model, we develop a prediction-driven phase-aware runtime optimization scheme that throttles concurrency so that power consumption can be reduced and performance can be set at the knee of the scalability curve of each program phase. The use of prediction reduces the overhead of searching the optimization space while achieving near-optimal performance and power savings. A thorough evaluation of our approach shows a reduction in power consumption of 10.8 percent, simultaneous with an improvement in performance of 17.9 percent, resulting in energy savings of 26.7 percent.

Interfacial Properties of LiTFSI and LiPF6-Based Electrolytes in Binary and Ternary Mixtures of Alkylcarbonates on Graphite Electrodes and Celgard Separator

For a better understanding of the adsorption behavior of alkylcarbonate-based electrolytes on graphite electrodes and Celgard separator for Li-ion batteries applications, the interface parameters are determined by contact angle and surface tension measurements. The correlation between these parameters and chemical compositions made of alkyl carbonate with a varying nature of lithium salts (LiPF6 and LiTFSI) and volume fractions of binary and ternary mixtures containing propylene carbonate (PC), ethylene carbonate (EC), and dimethyl carbonate (DMC) is investigated. From the obtained contact angle and surface tension (?L) values for each liquid, the dispersive and polar components of the surface tension (?Ld and ?Lp) of the electrolyte and interfacial free energy between the solid and liquid (?SL) were then calculated using the Young’s equation. The variation of contact angle (?) and the surface tension, as well as the work of adhesion (WA) of binary PC/DMC mixtures on PP, PE, and PET model surfaces were also measured and commented as function of volume fraction of PC in DMC. Finally, the Zisman’s critical surface tension (?C) for studied surfaces was then obtained showing positives slopes of cos ? versus ?L. This behavior is explained by a relative higher adsorption of alkylcarbonates to the hydrogenated supports or graphite. These results are decisive to understand the performance of electrolyte/electrode material/separator interfaces in lithium-ion battery devices.

Magnetic short-range order in the new ternary phase Mn8Pd15Si7

A new compound, Mn8Pd15Si7, is reported to crystallize in a face centered cubic unit cell of dimension a = 12.0141(2) angstrom, space groupFm (3) over barm, and can thus be classified as a G-phase. The crystal structure was studied by single crystal X-ray diffraction, X-ray and neutron powder diffraction and electron diffraction. A filled Mg6Cu16Si7 type structure was found, corresponding to the Sc11Ir4 type structure. The magnetic properties were investigated by magnetization measurements and Reverse Monte Carlo modeling of low temperature magnetic short-range order (SRO). Dominating near neighbor antiferromagnetic correlations were found between the Mn atoms and geometric frustration in combination with random magnetic interactions via metal sites with partial Mn occupancy were suggested to hinder formation of long-range magnetic order. 

Phase Equilibria of Binary and Ternary Systems Containing ILs, Dodecane, and Cyclohexanecarboxylic Acid

This paper reports both the binary and ternary phase behavior of ionic liquids for extracting cyclohexanecarboxylic acid (CCA) from dodecane. This system is a model for the extraction of acids representative of naphthenic acids found in crude oils. In order to develop an effective ternary liquid-liquid extraction system the preliminary selection of ionic liquids was based on CCA miscibility and the dodecane immiscibility with selected ILs. A wide range of ILs based on different cations, anions, cation alkyl-chain length, as well as the effect of temperature on the overall fluid phase behavior is reported. Factors such as variation of cation group, anion effect, alkyl-chain length, and temperature all impact the extraction to various degrees. The largest effects were found to be the lipophilicity of the IL cation and the co-ordination ability of the anion. While CCA capacity increased with lipophilicity of the cation, as did the dodecane. Highly coordinating anions such as trifluoroacetate and triflate demonstrated that highly efficient extraction could be obtained producing favorable tie-lines in the ternary phase diagram. Overall, this study demonstrates that ILs can selectively extract acids from hydrocarbon streams and offers possible treatment solutions for problems associated with the processing of high acid crude oils.

Performance Analysis of Randomized Distributed Space-Time Codes over Composite Gamma/Lognormal Fading Channels

This work investigates the end-to-end performance of randomized distributed space-time codes with complex Gaussian distribution, when employed in a wireless relay network. The relaying nodes are assumed to adopt a decode-and-forward strategy and transmissions are affected by small and large scale fading phenomena. Extremely tight, analytical approximations of the end-to-end symbol error probability and of the end-to-end outage probability are derived and successfully validated through Monte-Carlo simulation. For the high signal-to-noise ratio regime, a simple, closed-form expression for the symbol error probability is further provided.

Enabling Complexity-Performance Trade-Offs for Successive Cancellation Decoding of Polar Codes

Polar codes are one of the most recent advancements in coding theory and they have attracted significant interest. While they are provably capacity achieving over various channels, they have seen limited practical applications. Unfortunately, the successive nature of successive cancellation based decoders hinders fine-grained adaptation of the decoding complexity to design constraints and operating conditions. In this paper, we propose a systematic method for enabling complexity-performance trade-offs by constructing polar codes based on an optimization problem which minimizes the complexity under a suitably defined mutual information based performance constraint. Moreover, a low-complexity greedy algorithm is proposed in order to solve the optimization problem efficiently for very large code lengths.