Publication detail

An Allana Variance Computation from Very Large Datasets

KUNZ, J.

Original Title

An Allana Variance Computation from Very Large Datasets

English Title

An Allana Variance Computation from Very Large Datasets

Type

conference paper

Language

en

Original Abstract

This paper deals with calculating an Allan variance from a large datasets, which does not fit into a RAM memory. For this reason a parallel method is implemented in LabVIEW 2015 and tested on four different configurations of personal computer. An execution time was measured for a dataset of 2 754 448 000 samples. The time varies in dependence on calculated samples and the configuration of the computer. For the faster computers execution takes several minutes, whereas on the slower one the longest tested calculation of 100 samples per decade takes about two hours, which is still practical.

English abstract

This paper deals with calculating an Allan variance from a large datasets, which does not fit into a RAM memory. For this reason a parallel method is implemented in LabVIEW 2015 and tested on four different configurations of personal computer. An execution time was measured for a dataset of 2 754 448 000 samples. The time varies in dependence on calculated samples and the configuration of the computer. For the faster computers execution takes several minutes, whereas on the slower one the longest tested calculation of 100 samples per decade takes about two hours, which is still practical.

Keywords

Allan variance, large file, MEMS, gyroscope

Released

26.04.2018

Location

Brno

ISBN

978-80-214-5614-3

Book

Proceedings of the 24 th Conference STUDENT EEICT 2018

Pages from

363

Pages to

367

Pages count

5

Documents

BibTex


@inproceedings{BUT147184,
  author="Jan {Kunz}",
  title="An Allana Variance Computation from Very Large Datasets",
  annote="This paper deals with calculating an Allan variance from a large datasets, which does not
fit into a RAM memory. For this reason a parallel method is implemented in LabVIEW 2015 and
tested on four different configurations of personal computer. An execution time was measured for
a dataset of 2 754 448 000 samples. The time varies in dependence on calculated samples and the
configuration of the computer. For the faster computers execution takes several minutes, whereas on
the slower one the longest tested calculation of 100 samples per decade takes about two hours, which
is still practical.",
  booktitle="Proceedings of the 24 th  Conference STUDENT EEICT 2018",
  chapter="147184",
  howpublished="electronic, physical medium",
  year="2018",
  month="april",
  pages="363--367",
  type="conference paper"
}