Exploring energy-performance trade-offs for heterogeneous interconnect clustered VLIW processors

Clustered architecture processors are preferred for embedded systems because centralized register file architectures scale poorly in terms of clock rate, chip area, and power consumption. Although clustering helps by improving clock speed, reducing energy consumption of the logic, and making the design simpler, it introduces extra overheads by way of inter-cluster communication. This communication happens over long global wires which leads to delay in execution and significantly high energy consumption.In this paper, we propose a new instruction scheduling algorithm that exploits scheduling slacks of instructions and communication slacks of data values together to achieve better energy-performance trade-offs for clustered architectures with heterogeneous interconnect. Our instruction scheduling algorithm achieves 35% and 40% reduction in communication energy, whereas the overall energy-delay product improves by 4.5% and 6.5% respectively for 2 cluster and 4 cluster machines with marginal increase (1.6% and 1.1%) in execution time. Our test bed uses the Trimaran compiler infrastructure.

Construction and use of linear regression models for processor performance analysis

Processor architects have a challenging task of evaluating a large design space consisting of several interacting parameters and optimizations. In order to assist architects in making crucial design decisions, we build linear regression models that relate Processor performance to micro-architecture parameters, using simulation based experiments. We obtain good approximate models using an iterative process in which Akaike's information criteria is used to extract a good linear model from a small set of simulations, and limited further simulation is guided by the model using D-optimal experimental designs. The iterative process is repeated until desired error bounds are achieved. We used this procedure to establish the relationship of the CPI performance response to 26 key micro-architectural parameters using a detailed cycle-by-cycle superscalar processor simulator The resulting models provide a significance ordering on all micro-architectural parameters and their interactions, and explain the performance variations of micro-architectural techniques.

Improving Performance of Result Caches in Network Processors

Digest caches have been proposed as an effective method tospeed up packet classification in network processors. In this paper, weshow that the presence of a large number of small flows and a few largeflows in the Internet has an adverse impact on the performance of thesedigest caches. In the Internet, a few large flows transfer a majority ofthe packets whereas the contribution of several small flows to the totalnumber of packets transferred is small. In such a scenario, the LRUcache replacement policy, which gives maximum priority to the mostrecently accessed digest, tends to evict digests belonging to the few largeflows. We propose a new cache management algorithm called SaturatingPriority (SP) which aims at improving the performance of digest cachesin network processors by exploiting the disparity between the number offlows and the number of packets transferred. Our experimental resultsdemonstrate that SP performs better than the widely used LRU cachereplacement policy in size constrained caches. Further, we characterizethe misses experienced by flow identifiers in digest caches.

Adaptive Filtering Technique and DSP Based Implementation for High-Speed Distance Protection

The paper presents an adaptive Fourier filtering technique and a relaying scheme based on a combination of a digital band-pass filter along with a three-sample algorithm, for applications in high-speed numerical distance protection. To enhance the performance of above-mentioned technique, a high-speed fault detector has been used. MATLAB based simulation studies show that the adaptive Fourier filtering technique provides fast tripping for near faults and security for farther faults. The digital relaying scheme based on a combination of digital band-pass filter along with three-sample data window algorithm also provides accurate and high-speed detection of faults. The paper also proposes a high performance 16-bit fixed point DSP (Texas Instruments TMS320LF2407A) processor based hardware scheme suitable for implementation of the above techniques. To evaluate the performance of the proposed relaying scheme under steady state and transient conditions, PC based menu driven relay test procedures are developed using National Instruments LabVIEW software. The test signals are generated in real time using LabVIEW compatible analog output modules. The results obtained from the simulation studies as well as hardware implementations are also presented.

Extended histories: improving regularity and performance in correlation prefetchers

Data Prefetchers identify and make use of any regularity present in the history/training stream to predict future references and prefetch them into the cache. The training information used is typically the primary misses seen at a particular cache level, which is a filtered version of the accesses seen by the cache. In this work we demonstrate that extending the training information to include secondary misses and hits along with primary misses helps improve the performance of prefetchers. In addition to empirical evaluation, we use the information theoretic metric entropy, to quantify the regularity present in extended histories. Entropy measurements indicate that extended histories are more regular than the default primary miss only training stream. Entropy measurements also help corroborate our empirical findings. With extended histories, further benefits can be achieved by triggering prefetches during secondary misses also. In this paper we explore the design space of extended prefetch histories and alternative prefetch trigger points for delta correlation prefetchers. We observe that different prefetch schemes benefit to a different extent with extended histories and alternative trigger points. Also the best performing design point varies on a per-benchmark basis. To meet these requirements, we propose a simple adaptive scheme that identifies the best performing design point for a benchmark-prefetcher combination at runtime. In SPEC2000 benchmarks, using all the L2 accesses as history for prefetcher improves the performance in terms of both IPC and misses reduced over techniques that use only primary misses as history. The adaptive scheme improves the performance of CZone prefetcher over Baseline by 4.6% on an average. These performance gains are accompanied by a moderate reduction in the memory traffic requirements.

Optimization of HfO2 films for high transconductance back gated graphene transistors

Hafnium dioxide (HfO2) films, deposited using electron beam evaporation, are optimized for high performance back-gated graphene transistors. Bilayer graphene is identified on HfO2/Si substrate using optical microscope and subsequently confirmed with Raman spectroscopy. Back-gated graphene transistor, with 32 nm thick HfO2 gate dielectric, has been fabricated with very high transconductance value of 60 mu S. From the hysteresis of the current-voltage characteristics, we estimate the trap density in HfO2 to be in the mid 10(11)/cm(2) range, comparable to SiO2.

Zinc oxide nanostructures and high electron mobility nanocomposite thin film transistors

This paper reports on the synthesis of zinc oxide (ZnO) nanostructures and examines the performance of nanocomposite thin-film transistors (TFTs) fabricated using ZnO dispersed in both n- and p-type polymer host matrices. The ZnO nanostructures considered here comprise nanowires and tetrapods and were synthesized using vapor phase deposition techniques involving the carbothermal reduction of solid-phase zinc-containing compounds. Measurement results of nanocomposite TFTs based on dispersion of ZnO nanorods in an n-type organic semiconductor ([6, 6]-phenyl-C61-butyric acid methyl ester) show electron field-effect mobilities in the range 0.3-0.6 cm2V-1 s-1. representing an approximate enhancement by as much as a factor of 40 from the pristine state. The on/off current ratio of the nanocomposite TFTs approach 106 at saturation with off-currents on the order of 10 pA. The results presented here, although preliminary, show a highly promising enhancement for realization of high-performance solution-processable n-type organic TFTs. © 2008 IEEE.

High temperature direct double side cooled inverter module for hybrid electric vehicle application

In this paper a novel approach to the design and fabrication of a high temperature inverter module for hybrid electrical vehicles is presented. Firstly, SiC power electronic devices are considered in place of the conventional Si devices. Use of SiC raises the maximum practical operating junction temperature to well over 200°C, giving much greater thermal headroom between the chips and the coolant. In the first fabrication, a SiC Schottky barrier diode (SBD) replaces the Si pin diode and is paired with a Si-IGBT. Secondly, double-sided cooling is employed, in which the semiconductor chips are sandwiched between two substrate tiles. The tiles provide electrical connections to the top and the bottom of the chips, thus replacing the conventional wire bonded interconnect. Each tile assembly supports two IGBTs and two SBDs in a half-bridge configuration. Both sides of the assembly are cooled directly using a high-performance liquid impingement system. Specific features of the design ensure that thermo-mechanical stresses are controlled so as to achieve long thermal cycling life. A prototype 10 kW inverter module is described incorporating three half-bridge sandwich assemblies, gate drives, dc-link capacitance and two heat-exchangers. This achieves a volumetric power density of 30W/cm3.

First principles based multiparadigm modeling of electronic structures and dynamics

Electronic structures and dynamics are the key to linking the material composition and structure to functionality and performance.

An essential issue in developing semiconductor devices for photovoltaics is to design materials with optimal band gaps and relative positioning of band levels. Approximate DFT methods have been justified to predict band gaps from KS/GKS eigenvalues, but the accuracy is decisively dependent on the choice of XC functionals. We show here for CuInSe₂ and CuGaSe₂, the parent compounds of the promising CIGS solar cells, conventional LDA and GGA obtain gaps of 0.0-0.01 and 0.02-0.24 eV (versus experimental values of 1.04 and 1.67 eV), while the historically first global hybrid functional, B3PW91, is surprisingly the best, with band gaps of 1.07 and 1.58 eV. Furthermore, we show that for 27 related binary and ternary semiconductors, B3PW91 predicts gaps with a MAD of only 0.09 eV, which is substantially better than all modern hybrid functionals, including B3LYP (MAD of 0.19 eV) and screened hybrid functional HSE06 (MAD of 0.18 eV).

The laboratory performance of CIGS solar cells (> 20% efficiency) makes them promising candidate photovoltaic devices. However, there remains little understanding of how defects at the CIGS/CdS interface affect the band offsets and interfacial energies, and hence the performance of manufactured devices. To determine these relationships, we use the B3PW91 hybrid functional of DFT with the AEP method that we validate to provide very accurate descriptions of both band gaps and band offsets. This confirms the weak dependence of band offsets on surface orientation observed experimentally. We predict that the CBO of perfect CuInSe₂/CdS interface is large, 0.79 eV, which would dramatically degrade performance. Moreover we show that band gap widening induced by Ga adjusts only the VBO, and we find that Cd impurities do not significantly affect the CBO. Thus we show that Cu vacancies at the interface play the key role in enabling the tunability of CBO. We predict that Na further improves the CBO through electrostatically elevating the valence levels to decrease the CBO, explaining the observed essential role of Na for high performance. Moreover we find that K leads to a dramatic decrease in the CBO to 0.05 eV, much better than Na. We suggest that the efficiency of CIGS devices might be improved substantially by tuning the ratio of Na to K, with the improved phase stability of Na balancing phase instability from K. All these defects reduce interfacial stability slightly, but not significantly.

A number of exotic structures have been formed through high pressure chemistry, but applications have been hindered by difficulties in recovering the high pressure phase to ambient conditions (i.e., one atmosphere and room temperature). Here we use dispersion-corrected DFT (PBE-ulg flavor) to predict that above 60 GPa the most stable form of N₂O (the laughing gas in its molecular form) is a 1D polymer with an all-nitrogen backbone analogous to cis-polyacetylene in which alternate N are bonded (ionic covalent) to O. The analogous trans-polymer is only 0.03-0.10 eV/molecular unit less stable. Upon relaxation to ambient conditions both polymers relax below 14 GPa to the same stable non-planar trans-polymer, accompanied by possible electronic structure transitions. The predicted phonon spectrum and dissociation kinetics validate the stability of this trans-poly-NNO at ambient conditions, which has potential applications as a new type of conducting polymer with all-nitrogen chains and as a high-energy oxidizer for rocket propulsion. This work illustrates in silico materials discovery particularly in the realm of extreme conditions.

Modeling non-adiabatic electron dynamics has been a long-standing challenge for computational chemistry and materials science, and the eFF method presents a cost-efficient alternative. However, due to the deficiency of FSG representation, eFF is limited to low-Z elements with electrons of predominant s-character. To overcome this, we introduce a formal set of ECP extensions that enable accurate description of p-block elements. The extensions consist of a model representing the core electrons with the nucleus as a single pseudo particle represented by FSG, interacting with valence electrons through ECPs. We demonstrate and validate the ECP extensions for complex bonding structures, geometries, and energetics of systems with p-block character (C, O, Al, Si) and apply them to study materials under extreme mechanical loading conditions.

Despite its success, the eFF framework has some limitations, originated from both the design of Pauli potentials and the FSG representation. To overcome these, we develop a new framework of two-level hierarchy that is a more rigorous and accurate successor to the eFF method. The fundamental level, GHA-QM, is based on a new set of Pauli potentials that renders exact QM level of accuracy for any FSG represented electron systems. To achieve this, we start with using exactly derived energy expressions for the same spin electron pair, and fitting a simple functional form, inspired by DFT, against open singlet electron pair curves (H₂ systems). Symmetric and asymmetric scaling factors are then introduced at this level to recover the QM total energies of multiple electron pair systems from the sum of local interactions. To complement the imperfect FSG representation, the AMPERE extension is implemented, and aims at embedding the interactions associated with both the cusp condition and explicit nodal structures. The whole GHA-QM+AMPERE framework is tested on H element, and the preliminary results are promising.

Avaliação do copolímero de acrilonitrila e 2-vinil-2-oxazolina na eficiência de inibição de corrosão química em meio salino

Processos corrosivos geram perdas incalculáveis para a sociedade. Existe, portanto, uma incessante busca por métodos cada vez mais econômicos, eficientes e que não gerem impacto ao meio-ambiente. Polímeros heterocíclicos são macromoléculas de elevada performance que vêm sendo estudados como inibidores de corrosão por atenderem as premissas citadas acima. Neste trabalho realizou-se a modificação química da poliacrilonitrila com a incorporação de grupos 2-vinil-2-oxazolina no teor teórico de 20%. O copolímero obtido e a poliacrilonitrila não modificada foram caracterizados por espectroscopia na região do infravermelho e avaliados como inibidores de corrosão para alumínio e aço carbono em meio salino em temperatura de 60C. As concentrações utilizadas foram de 20, 50, 100 e 200 ppm e os períodos de checagem: 1, 2, 3 e 4 semanas. Os polímeros estudados apresentaram atividade inibitória de corrosão apenas para o aço carbono. Para o alumínio, o ensaio em branco apresentou a menor taxa de corrosão dentre todos os ensaios realizados

Avaliação in vitro da incorporação de clorexidina nas propriedades físico-químicas de resinas macias

A incorporação do diacetato de clorexidina em resinas macias para reembasamento de próteses pode se tornar uma boa alternativa na terapia com antimicrobianos usados para candidíase oral, uma vez que é independente da cooperação do paciente. Este trabalho teve como objetivo analisar a hipótese que a incorporação de diacetato de clorexidina em resinas macias não alteraria as suas propriedades físico-químicas. Foram analisadas duas resinas macias, uma à base de polimetilmetacrilato (PMMA) e outra à base de polietilmetacrilato (PEMA), com a incorporação de 0,5%, 1,0% e 2,0% de diacetato de clorexidina, além do grupo controle sem diacetato de clorexidina. Os espécimes foram armazenadas em água destilada a 37 durante 2hrs, 48 hrs, 7, 14, 21 e 28 dias e analisados quanto a cristalinidade do polímero através da difração de Raios X, a dureza mensurada com durômetro na escala shore A, o grau de conversão dos monômeros pela Espectroscopia de Infravermelho Transformada de Fourier (FTIR), a lixiviação de monômeros residuais e a liberação da clorexidina por Cromatografia Líquida de Alta Performace (HPLC). Os resultados mostraram que quanto à cristalinidade dos polímeros nas resinas macias à base de PMMA e de PEMA não alteraram após a incorporação de clorexidina, o que está relacionado à dureza Shore A, que aumentou ao longo do tempo, mas sem alteração significativa para nenhuma das concentrações de diacetato de clorexidina testadas. Já quanto ao grau de conversão, após a incorporação de clorexidina, a resina macia à base de PMMA não apresentou diferença estatisticamente significativa (р>0,05), porém a resina macia à base de PEMA apresentou diminuição significativa do grau de conversão (p<0,05), o que refletiu no aumento significativo da lixiviação de monômeros residuais nas concentrações 0,5% e 1,0% de clorexidina (p<0,05), principalmente nas primeiras 48 horas nessa resina. A resina macia à base de PMMA não apresentou alterações do padrão de lixiviação desses monômeros (р>0,05) e para ambas as resinas a cinética de liberação de clorexidina foi relacionada à lixiviação de monômeros residuais, que para a resina macia à base de PEMA foi maior significativamente nas primeiras 48 horas (p<0,05) e para a resina macia à base de PMMA não teve alteração estatisticamente significativa (р>0,05) se mantendo até o último dia de análise. Desse modo, conclui-se que a incorporação do diacetato de clorexidina não afetou a cristalinidade dos polímeros, inalterando a dureza nas duas resinas, porém o grau de conversão sofreu diminuição após a incorporação de diacetato de clorexidina na resina macia à base de PEMA, o que levou ao aumento da lixiviação de monômeros residuais e curta liberação de clorexidina; enquanto que a resina macia à base de PMMA, devido a não alteração significativa do seu grau de conversão, ocorreu menor lixiviação de monômeros residuais e liberação prolongada de clorexidina.

Large port count high-speed optical switch fabric for use within datacenters [Invited]

This paper reviews advances in the technology of integrated semiconductor optical amplifier based photonic switch fabrics, with particular emphasis on their suitability for high performance network switches for use within a datacenter. The key requirements for large port count optical switch fabrics are addressed noting the need for switches with substantial port counts. The design options for a 16×16 port photonic switch fabric architecture are discussed and the choice of a Clos-tree design is described. The control strategy, based on arbitration and scheduling, for an integrated switch fabric is explained. The detailed design and fabrication of the switch is followed by experimental characterization, showing net optical gain and operation at 10 Gb/s with bit error rates lower than 10-9. Finally improvements to the switch are suggested, which should result in 100 Gb/s per port operation at energy efficiencies of 3 pJ/bit. © 2011 Optical Society of America.

Binder free three-dimensional sulphur/few-layer graphene foam cathode with enhanced high-rate capability for rechargeable lithium sulphur batteries.

A novel ultra-lightweight three-dimensional (3-D) cathode system for lithium sulphur (Li-S) batteries has been synthesised by loading sulphur on to an interconnected 3-D network of few-layered graphene (FLG) via a sulphur solution infiltration method. A free-standing FLG monolithic network foam was formed as a negative of a Ni metallic foam template by CVD followed by etching away of Ni. The FLG foam offers excellent electrical conductivity, an appropriate hierarchical pore structure for containing the electro-active sulphur and facilitates rapid electron/ion transport. This cathode system does not require any additional binding agents, conductive additives or a separate metallic current collector thus decreasing the weight of the cathode by typically ∼20-30 wt%. A Li-S battery with the sulphur-FLG foam cathode shows good electrochemical stability and high rate discharge capacity retention for up to 400 discharge/charge cycles at a high current density of 3200 mA g(-1). Even after 400 cycles the capacity decay is only ∼0.064% per cycle relative to the early (e.g. the 5th cycle) discharge capacity, while yielding an average columbic efficiency of ∼96.2%. Our results indicate the potential suitability of graphene foam for efficient, ultra-light and high-performance batteries.

Preliminary study on the photoproduction of hydroxyl radicals in aqueous solution with Aldrich humic acid, algae and Fe(III) under high-pressure mercury lamp irradiation

Under a high-pressure mercury lamp (HPML) and using an exposure time of 4 h, the photoproduction of hydroxyl radicals ((OH)-O-.) could be induced in an aqueous solution containing humic acid (HA). Hydroxyl radicals were determined by high-performance liquid chromatography using benzene as a probe. The results showed that (OH)-O-. photoproduction increased from 1.80 to 2.74 muM by increasing the HA concentration from 10 to 40 mg L-1 at an exposure time of 4 h (pH 6.5). Hydroxyl radical photoproduction in aqueous solutions of HA containing algae was greater than that in the aqueous solutions of HA without algae. The photoproduction of (OH)-O-. in the HA solution with Fe(111) was greater than that of the solution without Fe(III) at pH ranging from 4.0 to 8.0. The photoproduction of (OH)-O-. in HA solution with algae with or without Fe(111) under a 250 W HPML was greater than that under a 125 W HPML. The photoproduction of (OH)-O-. in irradiated samples was influenced by the pH. The results showed that HPML exposure for 4 h in the 4-8 pH range led to the highest (OH)-O-. photoproduction at pH 4.0.

Subband electron properties of highly doped InAlAs/InGaAs metamorphic high-electron-mobility transistors on GaAs substrates

A Shubnikov-de Haas (SdH) oscillation measurement was performed on highly doped InAlAs/InGaAs metamorphic high-electron-mobility transistors on GaAs substrates at a temperature of 1.4 K. By analyzing the experimental data using fast Fourier transform, the electron densities and mobilities of more than one subband are obtained, and an obvious double-peak structure appears at high magnetic field in the Fourier spectrum. In comparing the results of SdH measurements, Hall measurements, and theoretical calculation, we found that this double-peak structure arises from spin splitting of the first-excited subband (i=1). Very close mobilities of 5859 and 5827 cm(2)/V s are deduced from this double-peak structure. The sum of the carrier concentration of all the subbands in the quantum well is only 3.95x10(12) cm(-2) due to incomplete transfer of the electrons from the Si delta -doped layer to the well. (C) 2001 American Institute of Physics.