Parallel implementation of octtree generation algorithm

Lal, P. Sojan; Unnikrishnan, A.; Jacob, K. Poulose

doi:10.1109/icip.1998.999084

Cited by 1 publication

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, since deploying multiple grid nodes on a single server allows decreasing the computation time, it is evident that, notwithstanding the improvements done with respect to the previous version of BiNS in terms of parallelism [3], the simulation has still a number of phases where the computation is serial, which slow down the execution. We think that we can cope with this issue by deploying multiple cubic subdomains within each grid node, by adopting an octtree structure to strongly parallelize the computation [38], possibly re-adapting the splitting during the simulation execution. Clearly, this would allow to run a grid node per server, thus effectively exploiting the available computing resources.…”

Section: Numerical Results For the Grid Implementation Of Binsmentioning

confidence: 99%

“…For ease of handling, the elementary volume used to split the overall environment is a cube, which is a common choice when a volume has to be split into different sub-volumes [38] to perform parallel computations, thanks to the symmetry of this geometrical figure.…”

Section: Distributed Implementation Of Bins2mentioning

confidence: 99%

“…differently from what can be done, for instance, using octree-based algorithms [38]. This choice does not imply degraded performance, but higher programming effort just for the simulation setup, since it is not fully automated.…”

Section: Distributed Implementation Of Bins2mentioning

confidence: 99%

“… the velocities of the particles before the collision and f v 1  and f v 2  those after the collision. We can introduce the momentum before and after collision: 38) This means that the final velocity in the system of the center of mass is a fraction of the initial velocity and propagates in the opposite direction. As a matter of fact, the direction depends on the incident angle of the particle on the plane which is tangent to considered surface in the point of collision with the particle.…”

Section: Appendix: Inelastic Collisionsmentioning

confidence: 99%

See 3 more Smart Citations

Simulation of molecular signaling in blood vessels: Software design and application to atherogenesis

Felicetti

Femminella

Reali

2013

Nano Communication Networks

View full text Add to dashboard Cite

This paper presents a software platform, named BiNS2, able to simulate diffusion-based molecular communications with drift inside blood vessels. The contribution of the paper is twofold. First a detailed description of the simulator is given, under the software engineering point of view, by highlighting the innovations and optimizations introduced. Their introduction into the previous version of the BiNS simulator was needed to provide to functions for simulating molecular signaling and communication potentials inside bounded spaces. The second contribution consists of the analysis, carried out by using BiNS2, of a specific communication process happening inside blood vessels, the atherogenesis, which is the initial phase of the formation of atherosclerotic plaques, due to the abnormal signaling between platelets and endothelium. From a communication point of view, platelets act as mobile transmitters, endothelial cells are fixed receivers, sticky to the vessel walls, and the transmitted signal is made of bursts of molecules emitted by platelets. The simulator allows evaluating the channel latency and the footprint on the vessel wall of the transmitted signal as a function of the transmitter distance from the vessels wall, the signal strength, and the receiver sensitivity. IntroductionThe capabilities of manipulating matter at the molecular scale has inspired a huge research effort for many years and has led to the design of sophisticated devices, commonly referred to as nanomachines. The potentials of these nano-devices span numerous areas [1][8][13], including medical science [2][14], environmental control, and material science. The technological progress has stimulated a lot of research focusing on different types of nanomachines, such as those related to the usage of carbon nanotubes and nanowires [51], and those making use of biological methods, which are the object of our analysis. Although full-fledged biological nanomachines do not yet exist, recently biochemists have done many progresses in this area, and now they are able to create at least functional cell components, such as artificial ribosomes, which can be used for the synthesis of proteins [49][50].However, the research regarding nanomachine networked coordination is still at an early stage.In the last years, some possible solutions for allowing nanomachines to exchange information have been proposed [1]. Clearly, achieving the objective of exchanging information at the nanoscale requires a deep exploration of the feasible mechanisms that allow designing the basic components of a communications system, such as an information encoder, a transmitter, a communication medium, a receiver, and an information decoder [13].Due to the heterogeneity of different environments and communication techniques that can be used at the nanoscale, it is unfeasible to identify general models, valid for most of nanocommunication systems. For example, signaling within a lymph node, or within blood vessels, or between brain cells, make use of mechanisms specific for each ...

show abstract

Section: Numerical Results For the Grid Implementation Of Binsmentioning

confidence: 99%

Section: Distributed Implementation Of Bins2mentioning

confidence: 99%

Section: Distributed Implementation Of Bins2mentioning

confidence: 99%

Section: Appendix: Inelastic Collisionsmentioning

confidence: 99%

See 2 more Smart Citations