Mesh-particle interpolations on graphics processing units and multicore central processing units

Particle-mesh interpolations are fundamental operations for particle-in-cell codes, as implemented in vortex methods, plasma dynamics and electrostatics simulations. In these simulations, the mesh is used to solve the field equations and the gradients of the fields are used in order to advance the particles. The time integration of particle trajectories is performed through an extensive resampling of the flow field at the particle locations. The computational performance of this resampling turns out to be limited by the memory bandwidth of the underlying computer architecture. We investigate how mesh-particle interpolation can be efficiently performed on graphics processing units (GPUs) and multicore central processing units (CPUs), and we present two implementation techniques. The single-precision results for the multicore CPU implementation show an acceleration of 45-70×, depending on system size, and an acceleration of 85-155× for the GPU implementation over an efficient single-threaded C++ implementation. In double precision, we observe a performance improvement of 30-40× for the multicore CPU implementation and 20-45× for the GPU implementation. With respect to the 16-threaded standard C++ implementation, the present CPU technique leads to a performance increase of roughly 2.8-3.7× in single precision and 1.7-2.4× in double precision, whereas the GPU technique leads to an improvement of 9× in single precision and 2.2-2.8× in double precision.     

Efficient algorithms for constructing broadcast disks programs in asymmetric communication environments

A well-known technique for broadcast program construction is the Broadcast Disks technique. However, in the Broadcast Disks approach there are some important disadvantages. For example some parts of the broadcast program remain empty during the construction procedure and the disk relative frequencies have to be selected very carefully. This paper gives a new approach of the broadcast program construction in a Broadcast Disk System, presenting three new algorithms, which face successfully these problems. According to this approach, the broadcast program is constructed with the minimum possible length, while retaining the selected disk relative frequencies and keeps the average delays on retrieving data-items low. The constructed broadcast programs have no empty parts, and retain their desired properties in any numbered combination of disk relative frequencies. We implemented all methods in a simulator application and experimental results show that our approach is more efficient than Broadcast Disks in all cases.     

The Mass Concentration and Elemental Composition of Indoor Aerosols in Suffolk and Onondaga Counties,New York

Indoor and outdoor aerosol sampling was conducted in two New York State Counties, Suffolk and Onondaga, during the period of January 6 and April 15, 1986. Week-long fine particle mass samples were collected indoors and outdoors for a total of 596 samples taken in 394 homes. The aerosol samples were analyzed by X-ray fluorescence for the following elements: Si, S, Cl, K, Ca, V, Mn, Fe, Ni, Cu, Zn, As, Se, Br, Cd, and Pb. Lastly, aerosol mass concentrations were determined gravimetrically. Homes included in this study had one or more of the following sources: cigarette smoking, kerosene heaters, wood burning and gas stoves. Homes with none of the above sources were also included. Comparisons between mass and elemental concentrations among the different home groups allowed the investigation of the elemental profiles and importance of the indoor sources. From the five investigated sources, cigarette smoking was found to be the most important, and to a lesser extent, kerosene heaters. Finally, gas stoves did not contribute to the observed indoor aerosol mass and elemental concentrations, but other unknown indoor sources did contribute to indoor aerosol concentrations. Room-to-room differences in mass and elemental concentrations were also investigated. The results of these analyses suggested that concentrations measured in the living area and kitchen are identical, within analytical errors. Therefore, the indoor environment can be characterized as well mixed.     

On the beam tilting and bit rate selection in mobile systems

Beam tilting of a mobile Base Station (BS), as a technique that limits the transmitted signals to each cell, introduces a way of increasing the system bit rate over a wider range than that without beam tilting. In this paper, the general approach of a BS antenna is being evaluated giving the system's response for different half power beamwidths (HPBW) and different types of propagation environments. The critical rate of the system versus the level of diffuse reflection, the communication distance, the excess distance and the height of the BS antenna are being examined.     

Durability of a sandstone used as a principal building and decorative material in ancient Abdera,Xanthi, N. Greece

Stones have been used as a construction material in Greece, since the 7th century B.C. Most of the public and other important buildings in ancient greek cities were made of stones. Various types of rocks have been used, from soft rock like travertine, soft sandstone, marl limestone, etc., to hard rocks like maarble, hard limestone, schist, etc. Some of these rocks, especially the soft ones, created problems of deterioration, through the time. One such a case, the deterioration of a soft sandstone used in Ancient Abdera, N. Greece, is examined in this paper. This sandstone is a molassic origin rock, used both as a building and decorative material since the antiquity. Sampling were made both from modern quarries (fresh rock) and the ancient settlement. The mineralogy, the granulometry, the physical and technical properties, the salt resistance and durability of the rock were examined in order to elucidate the possible causes of its deterioration in construction works. From the results obtained, it is concluded that the rock varies much in the field: in composition, in texture, and in coarseness. It consists of sand with remnants of rock fragments coming from the nearby metamorphic rocks of Rodope range. Petrographically the rock varies from lithic wacke (conglomerate, breaccia) (rock fragments >15%), to arkosic (feldspathic) wacke (feldspars>25%), to greywacke (clay>20%). According to its coarseness it varies from very coarse (mean grain size >4 mm) to very fine (grain size <0.2 mm). Its feldspars are in different stage of weathering, from intact to completely kaolinized. The matrix material, in most cases is more than 15% (>15%) and consists mainly of silica and clay in different proportions; ferrous oxides and/or hydroxides are present too. The CaCO₃ is usually low or non existant. The rock porosity is medium to high. The physical and technical properties of the rock in the field varies according to its mineralogy, texture (coarseness, matrix material, porosity), water content, etc. In general, the coarse grained members, with low clay content, low porosity and low water content are the most hard and durable. The water content places an important role in the strength of the rock mostly in clay reach members. The amount of matrix material, the porosity, and the strength of the rock decrease as its grain size (coarseness) increase. The main cause of the rock deterioration, in the construction sites is considered to be the water absorption and the salt crystallization (sea and/or ground water salts) near or under the rock surface. This is due to high clay content and the weak matrix material. Consequently, when the rock is far from the sea and/or from the foundation of the structure or in dry environments, it is quite durable with no problem of deterioration. Problems are created when the rock has been used in structures near the sea or in foundation. The possible remedial measures to be taken, in the ancient sites, are the protection of the rock from ground and/or sea water absorption and the related salts, by sheltering, water repelant coatings, and/or stabilizing agents.     

The non-genetic determinants of central adiposity

Central adiposity carries an increased risk of non-insulin dependent diabetes mellitus (NIDDM), cardiac disease, hypertension and death, and is closely related to insulin resistance. Genetic factors explain a large proportion of the population variance in central adiposity, although the genotypic characteristics remain obscure. Hormonal factors such as endogenous sex steroid levels, the menopause, hormone replacement therapy and cortisol may influence body fat partitioning. The link between dietary factors and central adiposity is controversial, with contradictory results in the literature. Smoking is associated with lower total body fat, but investigations of its influence on central adiposity have also yielded contradictory results. Higher levels of physical activity are associated with lesser amounts of central fat, both cross-sectionally and in intervention studies. Some of the contradictory results regarding putative influences on central adiposity may be due to limitations of some of the anthropometric parameters of central adiposity, such as the waist-hip ratio. Further research is required to clarify the relationships between many of these factors and with both compartments of central adiposity: subcutaneous abdominal and intraabdominal adipose tissue.     

A note on the homogeneity of the roughness on oblique finish turned surfaces

The homogeneity of the roughness on oblique finish turned surfaces is investigated statistically by an analysis of variance technique applied to experimental data obtained during the useful life of a carbide turning tool. In general, a highly significant row effect (due to tool wear) has been established. A significant column effect has also been found to be present in some of the selected specimens, particularly in those of smaller length. The results of this analysis provide a guide for surface roughness measurements in practice.     

Reduced-order performance of parallel and series-parallel identifiers with weakly observable parasitics

The stability properties of discrete-time parallel and series-parallel identifiers with respect to a specific model-plant order mismatch are analyzed. While in a deterministic environment with no modeling error the two schemes give identical results, when used in a deterministic environment with modeling error their performance is different. We assume a singularly perturbed state representation for the plant where the modeling error consists of fast parasitics which are weakly observable in the plant output. Detailed bounds on parameter and output estimate errors are established and the robustness of the adaptive identifiers is established by showing that the error bound goes to zero as the modeling error goes to zero, i.e. as the parasitics become infinitely fast. The dependence of this residual identification error on the input signal, the neglected parasitics, and the initial error conditions is shown to be crucial. The bounds indicate possibilities for reducing the error by a proper choice of the input signal.     

Two-dimensional ultrahigh-density X-ray optical memory

Most important aspect of nanotechnology applications in the information ultrahigh storage is the miniaturization of data carrier elements of the storage media with emphasis on the long-term stability. Proposed two-dimensional ultrahigh-density X-ray optical memory, named X-ROM, with long-term stability is an information carrier basically destined for digital data archiving. X-ROM is a semiconductor wafer, in which the high-reflectivity nanosized X-ray mirrors are embedded. Data are encoded due to certain positions of the mirrors. Ultrahigh-density data recording procedure can e.g., be performed via mask-less zone-plate-array lithography (ZPAL), spatial-phase-locked electron-beam lithography (SPLEBL), or focused ion-beam lithography (FIB). X-ROM manufactured by nanolithography technique is a write-once memory useful for terabit-scale memory applications, if the surface area of the smallest recording pits is less than 100 nm2. In this case the X-ROM surface-storage capacity of a square centimetre becomes by two orders of magnitude higher than the volumetric data density really achieved for three-dimensional optical data storage medium. Digital data read-out procedure from proposed X-ROM can e.g., be performed via glancing-angle incident X-ray micro beam (GIX) using the well-developed X-ray reflectometry technique. In presented theoretical paper the crystal-analyser operating like an image magnifier is added to the set-up of X-ROM data handling system for the purpose analogous to case of application the higher numerical aperture objective in optical data read-out system. We also propose the set-up of the X-ROM readout system based on more the one incident X-ray micro beam. Presented scheme of two-beam data handling system, which operates on two mutually perpendicular well-collimated monochromatic incident X-ray micro beams, essentially increases the reliability of the digital information read-out procedure. According the graphs of characteristic functions presented in paper, one may choose optimally the incident radiation wavelength, as well as the angle of incidence of X-ray micro beams, appropriate for proposed digital data read-out procedure.