This work presents a set of node-level optimizations to perform the assembly of edge finite element matrices that arise in 3D geophysical electromagnetic modelling on shared-memory architectures. Firstly, we describe ...
详细信息
ISBN:
(纸本)9781509014972
This work presents a set of node-level optimizations to perform the assembly of edge finite element matrices that arise in 3D geophysical electromagnetic modelling on shared-memory architectures. Firstly, we describe the traditional and sequential assembly approach. Secondly, we depict our vectorized and shared-memory strategy which does not require any low level instructions because it is based on an interpreted programming language, namely, Python. As a result, we obtained a simple parallel-vectorized algorithm whose runtime performance is considerably better than sequential version. The set of optimizations have been included to the work-flow of the Parallel Edge-based Tool for Geophysical Electromagnetic Modelling (PETGEM) which is developed as open-source at the Barcelona Supercomputing Center. Finally, we present numerical results for a set of tests in order to illustrate the performance of our strategy.
This study examines the performance of two axisymmetric nozzles which were designed to produce uniform, parallel flow with nominal Mach numbers of 4 and 8. A free-piston-driven shock tube was used to supply the nozzle...
This study examines the performance of two axisymmetric nozzles which were designed to produce uniform, parallel flow with nominal Mach numbers of 4 and 8. A free-piston-driven shock tube was used to supply the nozzle with high-temperature, high-pressure test gas. The inviscid design procedure treated the nozzle expansion in two stages. Close to the nozzle throat, the nozzle wall was specified as conical and the gas flow was treated as a quasi-one-dimensional chemically-reacting flow. At the end of the conical expansion, the gas was assumed to be calorically perfect and a contoured wall was designed (using Method-of-Characteristics) to convert the source flow into a uniform and parallel flow at the end of the nozzle. Performance was assessed by measuring Pitot pressures across the exit plane of the nozzles and, over the range of operating conditions examined, the nozzles produced satisfactory test flows. However, there were flow disturbances in the Mach 8 nozzle flow that persisted for significant times after flow initiation.
This paper describes use of ecommerce for easy crawling of information on an ecommerce website. As ecommerce has become an essential part in people's day-to-day activities of this modernized world. Searching &...
详细信息
ISBN:
(纸本)9781509002115
This paper describes use of ecommerce for easy crawling of information on an ecommerce website. As ecommerce has become an essential part in people's day-to-day activities of this modernized world. Searching & surfing data for ecommerce activity becomes tedious as lot many options & websites for single entity of ecommerce is presented before user. Internet is a huge set of database were many irrelevant data of no choice to particular task of user exist. In this paper we have designed a basic ontology of ecommerce along with its superclass & subclass as well as its siblings. This crawler design named ECOMMTOLOGY is an Ontology based Ecommerce application. Hence, we have classified ecommerce into hierarchy of business-to-business and business-to-consumer modules. The Ecommerce based ontology specifies each product details by searching, storing and retrieving user log details.
Deep neural models, particularly the LSTM-RNN model, have shown great potential for language identification (LID). However, the use of phonetic information has been largely overlooked by most existing neural LID metho...
详细信息
In this study, a novel feature selection framework is proposed to simultaneously perform classification and clinical scores prediction of Parkinson's disease (PD) via multi-modal neuroimaging data. Specifically, a...
详细信息
ISBN:
(纸本)9781509011735
In this study, a novel feature selection framework is proposed to simultaneously perform classification and clinical scores prediction of Parkinson's disease (PD) via multi-modal neuroimaging data. Specifically, a new feature selection model is devised to capture discriminative features to train support vector regression model for clinical scores (e.g., sleep scores and olfactory scores) prediction and support vector classification model for class label identification. Our method is evaluated on a public dataset of 208 subjects including 56 normal controls (NC), 123 PD and 29 scans without evidence of dopamine deficit (SWEDD) via a 10-fold cross-validation method. The experimental results demonstrate that multimodal data can effectively improve the performance in disease status identification and clinical scores prediction compared to one single modality. Our proposed method also outperforms the related methods.
Currently, cloud computing is facing different types of threats whether from inside or outside its environment. This may cause cloud to be crashed or at least unable to provide services to the requests made by clients...
详细信息
Currently, cloud computing is facing different types of threats whether from inside or outside its environment. This may cause cloud to be crashed or at least unable to provide services to the requests made by clients. In this paper, a new technique is proposed to make sure that the new node which asks to join the cloud is not composing a threat on the cloud environment. Our new technique checks the node before it will be guaranteed to join the cloud whether it runs malwares or software that could be used to launch an attack. In this way the cloud will allow only the clean node to join it, eliminating the risk of some types of threats that could be caused by infected nodes.
This paper describes an approach to identify suspected cybermob on social media. Many researches involve making predictions of group emotion on Internet (such as quantifying sentiment polarity), but this paper instead...
详细信息
This work reports a test structure to decide on the correctness of cache performance in chip multiprocessors (CMPs). The design targets the private L1 cache existing in Tiled CMPs architecture. It is developed around ...
详细信息
This work reports a test structure to decide on the correctness of cache performance in chip multiprocessors (CMPs). The design targets the private L1 cache existing in Tiled CMPs architecture. It is developed around the cellular automata (CA) structure proposed by von Neumann in 1950's. The theory of 3-neighborhood null-boundary CA is developed to record the inconsistent behavior of each of the processors L1 caches in CMPs. The special class of single length cycle attractor cellular automata accepts the (March) read/write status of cache word/line and evaluates the decision on the defective/inaccurate functioning of a cache module. This overcomes the inability of the classical design to identify defective behavior of CMPs cache. The test design further enables identification of the region of defective cache module in the CMPs that can help designers for defect diagnosis at design phase.
暂无评论