Software system of the Earth Simulator

The Earth Simulator (ES) is a large scale, distributed memory, parallel computer system consisting of 640 processor nodes (PN) with shared memory vector multi-processors (64GFLOPS/PN, 5120 APs in total, AP: arithmetic processor). All the nodes are connected via a high speed (16GB/s) single-stage crossbar network called the Interconnection Network (IN).

The operating system for the Earth Simulator is based on SUPER-UX, the UNIX operating system for the SX series scientific supercomputers. In order to realize high-performance parallel processing on the highly parallel machine, the operating system is enhanced for scalability.

The Earth Simulator system is managed as a two-level cluster system called the Super Cluster System. In the Super Cluster System, the Earth Simulator system is divided into 40 clusters (16 PNs/cluster). A single controller called Super Cluster Control Station (SCCS) manages all these clusters. This management system provides Single System Image (SSI) operation, management and job control for the large scale multi-node system.

The Job Scheduler (JS) and NQS running on the SCCS control all jobs of the system. They schedule the resources such as processing nodes and files which have not usually been treated as scheduling resources. This allows efficient scheduling of large scale jobs.

The MPI library (MPI/ES) and the HPF compiler (HPF/ES) are available for distributed parallel programming on the Earth Simulator. MPI/ES conforms to the MPI 2.0 standard and is optimized to exploit the hardware features. HPF/ES conforms to the core part of HPF 2.0 and supports some features of the HPF 2.0 approved extensions and HPF/JA 1.0 extensions. HPF/ES suitably handles the 3-level parallelism of the Earth Simulator system, that is, vectorization, shared-memory parallelization, and distributed-memory parallelization. Moreover, HPF/ES extends the language to easily handle irregular problems.     

Evaluation of Desulphurization Flux in CaO‐Al2O3‐BaO‐CeO2‐MgO System

The flux of the CaO‐Al₂0₃‐BaO‐CeO₂‐MgO system as a desulphurization flux containing no fluorine for the secondary metallurgy process was evaluated in this study. The flux composition was designed using the eutectic compositions of the binary systems. The melting and desulphurization abilities of the fluxes were evaluated by measuring their liquidus temperatures and the distribution ratios of sulphur between the fluxes and the carbon‐saturated iron or stainless steel. The lowest liquidus temperature of 1325°C was obtained by adding 5.7 mass% MgO to the 80mass%A‐20mass%B flux. (A: 12CaO‐7Al₂O₃, B: BaCeO₃+12mass%Al₂O₃). The distribution ratios of sulphur and sulphide capacities of the fluxes in this study were higher than those of the commercial product of calcium aluminate flux. This means that the CaO‐Al₂O₃‐BaO‐CeO₂‐MgO fluxes developed in this study have higher desulphurization and melting abilities compared with the commercial product of calcium aluminate flux.     

Electrochemical behaviors of uranium and plutonium at simultaneous recoveries into liquid cadmium cathodes

Experiments were conducted on simultaneous recovery of uranium and plutonium electrochemically into laboratory scale liquid cadmium cathodes (LCCs) at different U/Pu ratios in the salt phase, and the influence of the salt composition on the recovered amounts of uranium and plutonium, the morphologies of uranium and plutonium in the LCC, and the behavior of americium, which is present as a decay product of plutonium-241, were examined. As a result, it was shown that there is a threshold in the U/Pu ratio in the salt phase between 1/4.3 and 1/1.73 for the successful simultaneous recovery of uranium and plutonium up to 10 wt% in the LCC at high current efficiencies. In the LCC, uranium and plutonium existed in the forms of intermetallic compounds, (U, Pu)Cd₆ and (U, Pu)Cd₁₁, and also pure uranium metal. It was also revealed that americium associates with plutonium according to the separation factor during the LCC operation.     

Application of an ultrahigh-resolution scanning electron microscope (UHS-T1) to biological specimens

In 1985 we developed an ultrahigh-resolution scanning electron microscope with a resolution of 0.5 nm. It is equipped with a field emission gun and an objective lens with a very short focal length. In this study we report a survey of some different preparation techniques and biological specimens using the new scanning electron microscope. Intracellular structures such as cell organelles were observed surprisingly sharper than those observed by ordinary scanning electron microscopes. However, at magnifications over 250,000 X, platinum particles could be discerned as scattered pebbles on the surface of all structures in coated materials. Using an uncoated but conductively stained specimen, we successfully observed ribosomes on a rough endoplasmic reticulum at a direct magnification of 1 million. In these images some protrusions were recognized on the ribosomes. Ferritin and immunoglobulin G were used as samples of biological macromolecules. These samples were observed without metal coating and conductive staining. The ferritin particles appeared as rounded bodies without any substructure on the surface and immunoglobulin G as complexes of three-unit bodies. In the latter the central body might correspond to the Fc fragment and two side ones to Fab fragments. We assume that ultrahigh-resolution scanning electron microscopy is an effective means for observation of the cell fine structures and biological macromolecules. It will open a new research field in biomedicine.     

Elimination of Voids at Interface of <Emphasis Type="Italic">β</Emphasis>-SiC Films and Si Substrate by Laser CVD

Void-free β-SiC films were deposited on Si(001) substrates by laser chemical vapor deposition using hexamethyldisilane (HMDS) as the precursor. The effect of the time of introducing HMDS, i e, the substrate temperature when HMDS introduced (T_in), on the preferred orientation, surface microstructure and void was investigated. The orientation of the deposited SiC films changed from <001> to random to <111> with increasing T_in. The surface showed a layer-by-layer microstructure with voids above T_in ? 773 K, and then transformed into mosaic structure without voids at T_in= 298 K. The mechanism of the elimination of voids was discussed. At T_in =298 K, Si surface can be covered by an ultrathin SiC film, which inhibits the out-diffusion of Si atoms from substrate and prohibites the formation of the voids.     

Comparison of isolation methods using commercially available kits for obtaining extracellular vesicles from cow milk

Extracellular vesicles (EV) are important for delivering biologically active substances to facilitate cell-to-cell communication. Milk-derived EV are widely known because of their potential for immune enhancement. However, procedures for isolating milk-derived EV have not been fully established. To obtain pure milk-derived EV and accurately reveal their function, such procedures must be established. The aim of the present study was to compare methods using commercially available kits for isolating milk-derived EV. Initially, we investigated procedures to remove casein, which is the major obstacle in determining milk-derived EV purity. We separated whey using centrifugation only, acetic acid precipitation, and EDTA precipitation. Then, we isolated milk-derived EV by ultracentrifugation, membrane affinity column, size exclusion chromatography (SEC), polymer-based isolation, or phosphatidylserine-affinity isolation. Using EV count per milligram of protein, which is a good indicator of purity, we determined that acetic acid precipitation was the best method for removing casein. Using nanoparticle tracking analysis, protein quantity analysis, and RNA quantity analysis, we comprehensively compared each isolation method for its purity and yield. We found that SEC-based qEV column (Izon Science) could collect purer milk-derived EV at higher quantities. Thus, a combination of acetic acid precipitation and qEV can effectively isolate high amounts of pure extracellular vesicles from bovine milk.     

Dynamic Power Integrity Control of ATE for Eliminating Overkills and Underkills in Device Testing

This paper proposes a power integrity control technique for dynamically controlling power supply voltage fluctuations for a device under test (DUT), and demonstrates its effectiveness for eliminating the overkills/underkills due to the difference of power supply impedance between an automatic test equipment (ATE) and a practical operating environment of the DUT. The proposed method injects compensation currents into the power supply nodes on the ATE system in a feed-forward manner such that the ATE power supply waveform matches with the one on the customer’s operating environment of the DUT. A method for calculating the compensation current is also described. Experimental results show that the proposed method can emulate the power supply voltage waveform under a customer’s operating condition and eliminate 95 % of overkills/underkills in the maximum operating frequency testing with 105 real silicon devices. Limitations and applications of the proposed method are also discussed.     

Route guidance by a car navigation system based on augmented reality

Car navigation systems play a prominent role in road traffic safety and traffic regulation. However, it is necessary to improve the route guidance of car navigation systems so that they accurately and quickly recognize small differences in location. Thus, to increase the ease of understanding and safety of car navigation systems, navigation systems based on augmented reality have been proposed for providing guidance at road intersections. We are currently developing a car navigation system based on augmented reality, called AR‐Navi. We investigated designs for the display of road intersection guidance that can be easily understood even when limited information is available and proposed a “best shot” display method that does not use moving images. In addition, we implemented a prototype system that includes these methods and conducted driving experiments on public roads to evaluate the ease of understanding and safety of AR‐Navi. Using the evaluation results, we confirmed that the ease of understanding and safety is similar in the case of AR‐Navi and CG‐Navi. We also clarified the characteristics of AR‐Navi. © 2012 Wiley Periodicals, Inc. Electr Eng Jpn, 180(2): 43–54, 2012; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/eej.22278     

Gain and phase margin of optimal memoryless regulator of systems with time-delay

A class of linear quadratic regulators of the system with time-delay in the states is constructed by a memoryless feedback, and its robustness is evaluated. First, a static non-linear perturbation is inserted in the input channel, and a robust stability condition is derived. As its special case, it is shown that the regulator has infinite gain margin and at least 50% gain reduction tolerance. Next, a dynamic linear perturbation is inserted in the same point, and an alternative stability condition is derived. It assures that the regulator has at least 60o phase margin. These results coincide with those of the finite dimensional LQ regulator.     

Augmented reality-based block piling game with superimposed collapse prediction

Understanding what cannot be seen is difficult. Physical behavior can be explained on the basis of physical theories even if the behavior cannot be observed. Explanation of what is physically happening in the real world would become easy, however, if annotations were superimposed on the real objects. Herein, the authors demonstrate how an understanding of a physical event can be facilitated by overlapping a real-world situation with a simulation that predicts a future state. This idea is demonstrated in a game application in which a player stacks blocks into a pile until it collapses. In general, it is easy to estimate whether a block on the edge of a table will fall or not. However, it is more difficult to predict whether a stack of many blocks will collapse, and in what manner the stack will collapse. Even though previous research has demonstrated that the problem of how two-dimensionally stacked blocks collapse can be reduced to solving a sequence of convex quadratic programs, algorithms for convex quadratic programs require massive computational resources. Hence, the authors developed a fast and new algorithm based on a linear program. The proposed algorithm realizes real-time simulation based on physics that superimposes predicted collapse. The block that is predicted to fall is superimposed on the real block with a lit background projection. The system was evaluated in an experiment, and superimposed augmented reality annotation was observed to be efficient. The system was also demonstrated in game contests and received positive feedback and comments.