Number of the records: 1
Erratic server behavior detection using machine learning on streams of monitoring data
- 1.
SYSNO ASEP 0538411 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title Erratic server behavior detection using machine learning on streams of monitoring data Author(s) Adam, Martin (FZU-D) ORCID
Magnoni, L. (CH)
Pilát, M. (CZ)
Adamová, Dagmar (UJF-V) RID, ORCID, SAINumber of authors 4 Article number 07002 Source Title EPJ Web of Conferences, 245. - Les Ulis : EDP Sciences, 2020 / Doglioni C. ; Jackson P. ; Kamleh W. ; Kim D.Y. ; Silvestris L. ; Stewart G.A. - ISSN 2100-014X Pages s. 1-8 Number of pages 8 s. Publication form Online - E Action International Conference on Computing in High Energy and Nuclear Physics (CHEP 2019) /24./ Event date 04.11.2019 - 08.11.2019 VEvent location Adelaide Country AU - Australia Event type WRD Language eng - English Country FR - France Keywords machine learning ; monitoring Subject RIV JD - Computer Applications, Robotics OECD category Automation and control systems Subject RIV - cooperation Nuclear Physics Institute - Elementary Particles and High Energy Physics R&D Projects LM2015058 GA MŠMT - Ministry of Education, Youth and Sports (MEYS) EF16_013/0001404 GA MŠMT - Ministry of Education, Youth and Sports (MEYS) LM2018104 GA MŠMT - Ministry of Education, Youth and Sports (MEYS) Institutional support FZU-D - RVO:68378271 ; UJF-V - RVO:61389005 DOI 10.1051/epjconf/202024507002 Annotation With the explosion of the number of distributed applications, a new dynamic server environment emerged grouping servers into clusters, utilization of which depends on the current demand for the application. To provide reliable and smooth services it is crucial to detect and fix possible erratic behavior of individual servers in these clusters. Use of standard techniques for this purpose requires manual work and delivers sub-optimal results. Using only application agnostic monitoring metrics our machine learning based method analyzes the recent performance of the inspected server as well as the state of the rest of the cluster, thus checking not only the behavior of the single server, but the load on the whole distributed application as well. We have implemented our method in a Spark job running in the CERN MONIT infrastructure. In this contribution we present results of testing multiple machine learning algorithms and pre-processing techniques to identify the servers erratic behavior. Workplace Institute of Physics Contact Kristina Potocká, potocka@fzu.cz, Tel.: 220 318 579 Year of Publishing 2021
Number of the records: 1