Accurate and fast performance estimation methods for modern and future multi-core systems are the focal point of much research due to the complexity associated with such architectures. The communication architecture of such systems has a huge impact on the performance and power of the whole system. Architects need to explore many design possibilities by using performance estimation techniques at early stages of design to make design decisions earlier in the design cycle. While software developers need to develop and test applications for the target architecture and gather performance measurements as early in the design cycle as possible. Full system simulation techniques provide accurate performance values but are extremely time consuming. Static analysis techniques are fast but cannot capture the dynamic behavior associated with shared resource contention and arbitration. Moreover, synthetic traffic patterns have been used to analyze the communication architecture however, such patterns are not realistic enough. We propose a statistical based model to predict the dynamic cost of bus arbitration on the performance of a bus architecture. The proposed model uses workload trace of the actual applications and benchmarks to capture the real application traffic behavior. Statistics on the traffic patterns are collected and input to the analytical model which calculates performance values for the communication architecture under consideration. By knowing the performance measures, designers can avoid over and under-design of the communication architecture. This paper builds up on a previously developed performance estimation model. The previous work modeled single and burst bus-transfers, however only one interfering bus master at a time for each blocked bus request was considered. The proposed, improved accuracy model considers multiple interfering masters for each blocked request hence improving the estimation accuracy especially for traffic intensive applications and many PE architectures. Experiments are performed for two different architectures i.e., 4 processing elements connected via a shared bus and 8 processing elements connected via a shared bus. Results show no significant difference in accuracy compared to previously developed model, for low traffic applications SPARSE and ROBOT however notable accuracy improvement for traffic intensive applications. Maximum estimation error is reduced from 1.75% to 0.6% for FPPPP and from maximum 13.91% to 8.8% for FFT on the 4PE architecture. On the 8PE architecture, maximum estimation error is reduced from 11.8% to 2.7% for the FPPP benchmark. Moreover simulation speed-up for the proposed technique over simulation method is reported.

抄録全体を表示

The progress of the Cryptomeria japonica pollen front for 2017 (Fig 1) was monitored by tracking the first day of pollen release. A Durham sampler was used at 77 sites, and an IS-Rotary sampler at 8 sites distributed from Kyushu to Hokkaido. Results from the Durham sampler showed that, by early February, the pollen front had reached the western part of Chugoku and southern parts of Kanto and Tokai. By mid February, the front had moved up to the southern and northern areas of Kanto, Tokai and the southern area of Tohoku, the southern to central parts of Kinki, and the whole of Chugoku and Shikoku, with the exception of its southern region. By late February the front had moved up to the central area of Tohoku, central Chubu and the whole of Hokuriku and northern Kinki. By early March, the front was recorded in central to northern parts of Tohoku and, unexpectedly, to central Tokai. While by mid March the front had reached the northern part of Tohoku district. The front finally arrived in the eastern and middle parts of Hokkaido by early April.

抄録全体を表示

The objective of this study was to increase the formation rate of embryos from suspension cultures of Taranoki (Aralia elata). The suspension cultures were prepared from leaf-derived calli. Of the various culture durations, 30 days of culture yielded the best growth of cell suspension cultures in a BT liquid medium supplemented with 2,4-D. Somatic embryogenesis has never been observed without the addition of exogenous plant growth regulators (PGRs). The highest percentage of somatic embryo formation (93.3%) was observed in a BT liquid medium supplemented with 1.0 μM BAP+5.0 μM 2,4-D. The PGR-free semi-solid BT medium showed the highest regeneration frequency of cotyledonary-shaped embryos (100%).

抄録全体を表示

本研究では, 減圧加熱下における溶融飛灰からの重金属塩化物の揮発特性を, 実溶融飛灰および模擬飛灰を用いて実験的に検討した。
その結果, 減圧加熱下における単一重金属塩化物 (ZnCl₂, PbCl₂, CuCl) の揮発速度は, 減圧度の増大に伴って増加した。一方, ZnCl₂-PbCl₂-CuCl混合物からの各重金属塩化物の揮発速度は, 単一重金属塩化物のそれよりも小さくなり, 試料中重金属の含有量の減少に伴って減少した。このときの重金属塩化物の揮発速度は, 試料中の重金属濃度の1次式で表された。さらに減圧加熱下における重金属塩化物の揮発挙動に及ぼす灰組成の影響を調べた結果, ZnCl₂揮発量は, Al₂O₃含有量の増加に伴って減少した。また, NaCl, KCl, CaCl₂の共存によって, 溶融飛灰中重金属の揮発量は, NaCl, KClおよびCaCl₂を含有しない模擬飛灰に対して0.27～0.65倍となることが認められた。

抄録全体を表示

This paper reviews our simulation study by numerical calculation based on two-, three- and multi-grain-size models on WC grain growth in fine grained WC-Co hardmetal doped with VC. The study aimed to presume or predict the following: (1) the cause and conditions for the abnormal grain growth which occurs in some cases in the fine grained hardmetal prepared from fine WC powders with mean grain size below about 0.2μm, and (2) how the mean grain size of the hardmetal varies with decreasing mean grain size of the WC starting powder to 0.1 μm or nano-meter size. The calculation results by these three kinds of models suggested the following, respectively: (1) the occurrence of the abnormal grain growth is generally substantial for WC starting powder with bimodal size-distribution and mean grain size below 0.1-0.2μm, and not due to the non-uniform distribution of the grain growth inhibitor, (2) the introduction of middle grains caused the disappearance of small (fine) grains during sintering, leading to abruptly increase of the mean grain size of alloy, and (3) the mean grain size of alloy becomes so large as about 0.3-0.4μm, even when the WC starting powders with nano-size as well as 0.1μm are used.

抄録全体を表示

This study proposes the use of multiobjective optimization of a conceptual hydrological model with perturbed data as a sampling method to reproduce the posterior distribution of parameters for the quantification of uncertainty. The Pareto front is found to be sensitive to perturbed data, so model parameters are optimized with different combinations of perturbed data sets to sample behavioral parameters. Latin Hypercube Sampling (LHS) method was used to sample the behavioral parameters in order to evaluate the performance of the proposed method. The performance of simulation for all parameter sets sampled by both methods is evaluated and presented in objective space. The proposed method sampled large parameter sets more efficiently near optimal compared with LHS. The study demonstrates that the Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II) a multiobjective optimizer, with perturbed data set can efficiently explore near optimal parameter space of a conceptual hydrological model.

抄録全体を表示

A 12-bit 1.25MS/s cyclic analog-to-digital converter (ADC) is designed and fabricated in 90nm CMOS technology, and only occupies an active area as small as 0.037mm². The proposed ADC is composed of a non-binary AD convertion stage, and a on-chip non-binary-to-binary digital block includes a built-in radix-value self-estimation scheme. Therefore, althouh a non-binary convertion architechture is adopted, the proposed ADC is the same as other stand-alone binary ADCs. The redundancy of non-binary 1-bit/step architecture relaxes the accuracy requirement on analog components of ADC. As a result, the implementation of analog circuits such as amplifier and comparator becomes simple, and high-density Metal-Oxide-Metal (MOM) capacitors can be used to achieve a small chip area. Furthermore, the novel radix-value self-estimation technique can be realized by only simple logic circuits without any extra analog input, so that the total active area of ADC is dramatically reduced. The prototype ADC achieves a measured peak signal-to-noise-and-distortion-ratio (SNDR) of 62.3dB using a poor DC gain amplifier as low as 45dB and MOM capacitors without any careful layout techniques to improve the capacitor matching. The proposed ADC dissipated 490µW in analog circuits at 1.4V power supply and 1.25Msps (20MHz clocking). The measured DNL is +0.94/-0.71LSB and INL is +1.9/-1.2LSB at 30kHz sinusoidal input.

抄録全体を表示

A practical form of the joint distribution of wave heights and periods is proposed in this paper. It is traditionally inevitable to employ the approximation of local-height and local-period in sense of simplicity. The ambiguous relation in the previous studies is pointed out, which is the doubtful assumption between the angular velocity and real-period, and then an altanative relation is introduced newly. It is an advantage in this study that the marginal distribution of wave height is obtained as the Rayleigh distribution. The proposed joint distribution is so adaptable that the effect of nonlinearity and wave-breaking will be easily taken into account.

抄録全体を表示

本研究では自然観察法を用い、1歳6ヶ月から2歳2ヶ月の23人の中国人幼児を観察し、その物を巡るやりとりが記録された。それらのやりとりが順位制のように順位によって構造化されているかどうかを明らかにするためにデータが分析された。その結果1)順位のマトリックスの強さも直線性も非常に低い。2)マトリックス上の順位と、担任による子どもの強さやリーダシップの評価値の相関は有意ではない。3)子どもの優劣関係は直線的ではなく錯綜している。ということが明らかになった。全体として所有に関してこれらの幼児集団は順位制によって組織化されていないことが示された。

抄録全体を表示

This paper intended to explore the problem solving behavior (PSB) of people in design tasks through simulated design process for interior works using interactive genetic algorithm (IGA). Through analysis of people's problem solving process when evaluating the juxtaposed images, it was revealed that people tend to do what they are certain of firstly, and make harder decisions later. The strategy helps them to unfold the problem gradually. It was also found that people did not tend to move their eyes to a faraway image in the interface constantly, which is more convenient for them.

抄録全体を表示

In order to evaluate quantitatively the initial mean grain size (do) and the grain size distribution (σo) for the occurrences of the considerable grain growth and abnormal grain microstructure observed in fine grained hardmetals, the grain growth was simulated by numerical calculation based on multi-grain size model where the rate-determining step of Ostwald ripening is the interface reaction (precipitation), following the previous studies based on 2- and 3- grain size models.
The results obtained were as follows: (1) The following three equations were derived for the growth of grains except for the smallest grain, the extinction of the smallest grain, and the mean grain size, respectively.
d_{i, t+Δt}=d_{i, t}+(81/32)K(1/d_{min, t}-1/d_{i, t})Δtd_{min, t+Δt}-(81/32)⋅K⋅(1/d_{min, t}²F_{min, t})[NOΣi=min+{d_{i, t}²F_{i, t}(1/d_{min, t}-1/d_{i, t})}]⋅Δt-d_t=NOΣi=min(F_{i, t}⋅d_{i, t})
where K is (32/81)4(2C_oσKVm²/vRT) and is the same as K in well-known equation of d₁²-d₀²=Kt. The value of K measured for typical fine grained WC-0.5mass%VC-10mass%Co hardmetal is 2.18×10^-2μm²/ks in the case of sintering at 1673K for 0-10.8ks.
(2) In the case of monomodal (normal) distribution of WC grain size, the abnormal grain microstructure dose not occur even at d₀ less than 0.2-0.lμm for the sintering at 1673K for 3.6ks. However, the considerable grain growth occurs and the mean grain size (d_3.6) of the sintered alloy becomes above about 0.4μm, irrespective of the value of σ_o. The result is similar to that obtained by 3-grain size model. (3) In the case of bimodal distribution consisting two kinds of normal distributions, the abnormal grain microstructure occurs when the total number of grains in the second distribution is extremely smaller than that in the first distribution and also the initial ratio of the mean grain size (d₀^2m) in the second distribution to that in the first distribution (d₀^1m) is above about 9, for example in the case of d₀^1m(≅d₀)=0.1μm. The d_3.6 is nearly the same as that in the monomodal distribution. The result is similar to that obtained by 2-grain size model. (4) From these results, it was suggested that the fine grained hardmetal with mean grain size less than 0.3μm is substantially difficult to be fabricated by controlling the initial distribution of WC powder as well as the initial mean grain size, as far as VC (5mass% to Co) is used as the grain growth inhibitor.

抄録全体を表示

This paper presents a method to solve the Bezoutian equation for the indirect pole-placement control of single-input single-output linear non-minimum phase systems with unknown constant parameters. The method consists of three steps; (i) the elementary transformation of the Sylvester resultant matrix into the upper triangular form by pivot searches, (ii) the exclusion of the common factors from each rows of the matrix obtained in the step (i) when a pivot becomes too small or zero, and (iii) the calculation of controller parameters from the linear simultaneous equations of which coefficient matrix is the transpose of the matrix obtained by the step (ii). Since the polynominal composed of the common factors is obtained directly from the row elements above the small pivot, the factrizations of the system polynomials are not required. These solusion procedures are illustrated by the numerical examples, and the validity of the proposed method is proved by using the concept of the linear transformation of state variables.

抄録全体を表示

居住者が発生する換気対象の室内における流動には代謝熱により発生する自然対流が関わる。強制換気の行われる部屋におけるこの影響を調べるために水槽モデルによる実験を行った。1〜2名で使用する小部屋の縮尺1/5の水槽と円柱形発熱モデルを使用し、グラスホフ数とレイノルズ数を相似則により合せて実験を行った。実験結果の検討のために熱直接排除率と給水バイパス率を導入した。天井付き排気口と、居住者との位置関係による換気効率の変化を実験した。自然対流が直接排気口に到達し、かつ排気口がこれを吸い込みやすい形状の場合、熱直接排除率は大きかったが、排気口と発熱モデルの平面上の位置がずれている場合や、排気口前面に拡散板がある場合にはこの率の差は10%程度であった。

抄録全体を表示

Five separate fractions of noble gases were extracted and analyzed as a 150g sample of the Canyon Diablo iron meteorite was heated to successively higher temperatures by direct coupling with an RF coil. Isotopic compositions of the light gases, He, Ne and Ar, demonstrate that essentially their entire inventory was produced by spallation reactions. Spallation products are also a prominent feature of the isotopes of Kr and Xe, but in each gas fraction the spallogenic heavy gases were mixed with a nonspallogenic component. The isotopic composition of Kr and Xe trapped in Canyon Diablo are shown to be type-Y, rather than the type-X Kr and Xe that is found with planetary He and Ne in mildly oxidized carbon fractions of chondrites. Large variations are observed in ratios of the light spallogenic noble gases. Spallation products in Kr and Xe tend to correlate with those in Ne, rather than in Ar, suggesting that diffusive fractionation is not responsible for variations in the ²¹Ne/³⁸Ar ratio. It is suggested that portions of the material in Canyon Diablo may have been preirradiated. Xenon contains a large component of excess ¹²⁹Xe and ¹³¹Xe from neutron capture on Te at energies above 0.4ev, but there is no indication that the decay of primordial ¹²⁹I or the incorporation of exotic interstellar dust produced the excess ¹²⁹Xe or ¹³¹Xe in our sample of Canyon Diablo. The maximum amount of excess ¹³⁶Xe in our sample from fission of the hypothetical superheavy elements is shown to be 600atoms/g if the trapped Xe is atmospheric in composition; however, if the trapped Xe is AVCC, it might be as high as 6, 000atoms/g.

抄録全体を表示