Topic
Spark (mathematics)
About: Spark (mathematics) is a research topic. Over the lifetime, 7304 publications have been published within this topic receiving 63322 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: A fundamental study of electro discharge machining (EDM) based on the physics of an arc and heat transfer theory has been carried out in this paper, where the field equations for electric potential and temperature in the spark region are simultaneously solved by employing the finite element method.
Abstract: A fundamental study of electro discharge machining (EDM) based on the physics of an arc and heat transfer theory has been carried out. The field equations for electric potential and temperature in the spark region are simultaneously solved by employing the finite element method. Using the criterion of constant current at any cross section of a spark, the arc radii at different cross sections are corrected until convergence. The final spark shape obtained is noncylindrical, and has different radii at different cross sections. Also, the percent of heat input absorbed by cathode, anode, and dielectric has been calculated. The computed relative electrode wear has been compared with experimental results.
125 citations
TL;DR: In this article, a number of recent developments in spark production of nanoparticles that are important for the synthesis of nanopowders and nanoparticulate materials are discussed, including recent improvements and theoretical and practical experience in controlling the main particle parameters determining the product properties.
124 citations
TL;DR: GeoSpark is presented, which extends the core engine of Apache Spark and SparkSQL to support spatial data types, indexes, and geometrical operations at scale and achieves up to two orders of magnitude faster run time performance than existing Hadoop-based systems.
Abstract: The paper presents the details of designing and developing GeoSpark, which extends the core engine of Apache Spark and SparkSQL to support spatial data types, indexes, and geometrical operations at scale. The paper also gives a detailed analysis of the technical challenges and opportunities of extending Apache Spark to support state-of-the-art spatial data partitioning techniques: uniform grid, R-tree, Quad-Tree, and KDB-Tree. The paper also shows how building local spatial indexes, e.g., R-Tree or Quad-Tree, on each Spark data partition can speed up the local computation and hence decrease the overall runtime of the spatial analytics program. Furthermore, the paper introduces a comprehensive experiment analysis that surveys and experimentally evaluates the performance of running de-facto spatial operations like spatial range, spatial K-Nearest Neighbors (KNN), and spatial join queries in the Apache Spark ecosystem. Extensive experiments on real spatial datasets show that GeoSpark achieves up to two orders of magnitude faster run time performance than existing Hadoop-based systems and up to an order of magnitude faster performance than Spark-based systems.
124 citations
TL;DR: Two of the comparison of - Hadoop Map Reduce and the recently introduced Apache Spark - both of which provide a processing model for analyzing big data are discussed, both of whom vary significantly based on the use case under implementation.
Abstract: Data has long been the topic of fascination for Computer Science enthusiasts around the world, and has gained even more prominence in the recent times with the continuous explosion of data resulting from the likes of social media and the quest for tech giants to gain access to deeper analysis of their data This paper discusses two of the comparison of - Hadoop Map Reduce and the recently introduced Apache Spark - both of which provide a processing model for analyzing big data Although both of these options are based on the concept of Big Data, their performance varies significantly based on the use case under implementation This is what makes these two options worthy of analysis with respect to their variability and variety in the dynamic field of Big Data In this paper we compare these two frameworks along with providing the performance analysis using a standard machine learning algorithm for clustering (K- Means)
124 citations
01 Jan 2018
TL;DR: Recurrent Neural Networks based Models: Long-Short-Term-Memory (LSTM) and Gated-Recurrent-Unit (GRU) and GRU-RNN are dealt with to deal with electric load forecasting challenge.
Abstract: Electric load forecasting has a significant role in power grids in order to facilitate the decision making process of energy generation & consumption. Long term forecasting is not feasible as there might be an uncertainty in the prediction because of irregular increase in the demand for power with the growing population and dependency on electric power. Since the behaviour of electric load time series is very much non-linear and seasonal, Neural Networks are best suited model for learning the Non-Linear behaviour within the data and for forecasting purpose. This paper deals with the Recurrent Neural Networks based Models: Long-Short-Term-Memory (LSTM) and Gated-Recurrent-Unit (GRU) to deal with this challenge. Observations have been made based on the distributed implementation of various configurations of LSTM-RNN & GRU-RNN on spark clusters for hyper parameter tuning purpose and deploying best suited configuration with least RMSE value using apache memos resource management.
121 citations