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: In this paper, a series of experiments executed to characterize the physical and chemical nature of spark discharge in argon at atmospheric pressure are described. Butts et al. investigated the expansion properties of the spark and the distribution and longevity of cathode material in and around the spark channel following cessation of the discharge current.
16 citations
14 Apr 2015
16 citations
TL;DR: In this paper, the necessary conditions for the satisfactory parallel operation of spark gaps, with no transit time isolation between them, were deduced theoretically, and a 60 kv pressurized field-distortion spark with an inductance of 15-18 nH was described, and the influence of electrode arrangement and trigger circuit parameters on breakdown time and jitter was investigated.
Abstract: The necessary conditions for the satisfactory parallel operation of spark gaps, with no transit time isolation between them, are deduced theoretically. Such a capacitor bank with closely connected parallel spark gaps has been operated successfully. The development of a 60 kv pressurized field-distortion spark with an inductance of 15-18 nH is described, and the influence of electrode arrangement and trigger circuit parameters on breakdown time and jitter has been investigated.
16 citations
01 Nov 2015
TL;DR: A design scheme based on distributed architecture, mainly using Kafka, Storm and Spark clusters are proposed, and the real-time transmission among the structured data, unstructured data and database data are implemented.
Abstract: For the problem about the low shared efficiency and the poor integrating degree of the massive multivariate data, in the paper, we propose a design scheme based on distributed architecture, mainly using Kafka, Storm and Spark clusters, and the real-time transmission among the structured data, unstructured data and database data are implemented And the real-time analysis system of real-time data and statistical analysis system of offline data are developed Finally, the test results show the performance of Big Data processing platform, and demonstrate fully the advantage of the distributed architecture
16 citations
20 Jul 2016
TL;DR: Hylas is described, a tool for automatically optimising Spark queries embedded in source code via the application of semantics-preserving transformations, inspired by functional programming techniques of "deforestation", which eliminate intermediate data structures from a computation.
Abstract: Apache Spark is a popular framework for large-scale data analytics. Unfortunately, Spark's performance can be difficult to optimise, since queries freely expressed in source code are not amenable to traditional optimisation techniques. This article describes Hylas, a tool for automatically optimising Spark queries embedded in source code via the application of semantics-preserving transformations. The transformation method is inspired by functional programming techniques of "deforestation", which eliminate intermediate data structures from a computation. This contrasts with approaches defined entirely within structured query formats such as Spark SQL. Hylas can identify certain computationally expensive operations and ensure that performing them creates no superfluous data structures. This optimisation leads to significant improvements in execution time, with over 10,000 times improvement observed in some cases.
16 citations