GATE: improving the computational efficiency
نویسندگان
چکیده
GATE is a software dedicated to Monte Carlo simulations in Single Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET). An important disadvantage of those simulations is the fundamental burden of computation time. This manuscript describes three different techniques in order to improve the efficiency of those simulations. Firstly, the implementation of variance reduction techniques (VRTs), more specifically the incorporation of geometrical importance sampling, is discussed. A relative figure of merit was calculated for a standard setup, showing an efficiency enhancement of 5 − 15 by using this technique. After this, the newly designed cluster version of the GATE software is described. The experiments have shown that GATE simulations scale very well on a cluster of homogeneous computers for the case of low sensitivity (SPECT) setups and that an optimum can be derived for high sensitivity (PET) experiments using this cluster package of the GATE software. Finally, an elaboration on the deployment of GATE on the EGEE (Enabling Grids for E-Science in Europe) Grid will conclude the description of efficiency enhancement efforts. The latter has shown to be efficient but depending on the queuing policy of the site accepting the jobs. The three aforementioned methods improve the efficiency of GATE to a large extent and make realistic patient-specific overnight Monte Carlo simulations achievable.
منابع مشابه
IMPROVING COMPUTATIONAL EFFICIENCY OF PARTICLE SWARM OPTIMIZATION FOR OPTIMAL STRUCTURAL DESIGN
This paper attempts to improve the computational efficiency of the well known particle swarm optimization (PSO) algorithm for tackling discrete sizing optimization problems of steel frame structures. It is generally known that, in structural design optimization applications, PSO entails enormously time-consuming structural analyses to locate an optimum solution. Hence, in the present study it i...
متن کاملA A Survey of Methods For Analyzing and Improving GPU Energy Efficiency
Recent years have witnessed a phenomenal growth in the computational capabilities and applications of GPUs. However, this trend has also led to dramatic increase in their power consumption. This paper surveys research works on analyzing and improving energy efficiency of GPUs. It also provides a classification of these techniques on the basis of their main research idea. Further, it attempts to...
متن کاملSmartCell: A Coarse-Grained Reconfigurable Architecture for High Performance and Low Power Embedded Computing
This paper describes a novel coarse-grained reconfigurable architecture, namely SmartCell, targeted for high data throughput and computationally intensive applications. By integrating a large number of computational units with reconfigurable interconnection fabrics, SmartCell is able to provide stream processing capacity to achieve both performance and power efficiency. The preliminary results ...
متن کاملLattice Recurrent Unit: Improving Convergence and Statistical Efficiency for Sequence Modeling
Recurrent neural networks have shown remarkable success in modeling sequences. However low resource situations still adversely affect the generalizability of these models. We introduce a new family of models, called Lattice Recurrent Units (LRU), to address the challenge of learning deep multi-layer recurrent models with limited resources. LRU models achieve this goal by creating distinct (but ...
متن کاملA Computational Study on the Performance of Graphene Nanoribbon Field Effect Transistor
Despite the simplicity of the hexagonal graphene structure formed by carbon atoms, the electronic behavior shows fascinating properties, giving high expectation for the possible applications of graphene in the field. The Graphene Nano-Ribbon Field Effect Transistor (GNRFET) is an emerging technology that received much attention in recent years. In this paper, we investigate the device performan...
متن کامل