Skip to main navigation Skip to search Skip to main content

The Data Analytics Framework for XDMoD

  • SUNY Buffalo

Research output: Contribution to journalArticlepeer-review

4 Scopus citations

Abstract

Open XDMoD is an established, web-based software tool that facilitates monitoring of cyberinfrastructure (CI); including metrics on compute job performance, cloud usage, storage, science gateways, users, institutions, allocations, awards, and more. Centers can install Open XDMoD and populate it with their local CI information. Additionally, information is acquired daily for CI systems allocated by the National Science Foundation (NSF) Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS) program and placed in an extensive historical data warehouse as part of an ACCESS-hosted version of Open XDMoD. A user portal displays customizable charts that can be filtered and drilled down to provide a wide variety of reports quickly and easily. However, there will always be desired analyses for which the data are available but which cannot be performed because of limitations in the user interface. To remedy this situation, we have developed a Data Analytics Framework for XDMoD that provides an application programming interface (API) to the Open XDMoD data warehouse; enabling CI studies, user studies, return on investment studies, and any other studies that can utilize the data. The Data Analytics Framework provides data scientists the capability to access data from Open XDMoD and perform any desired analyses using analytics tools of their choice. The framework is available both for the ACCESS-hosted version of Open XDMoD for broad-ranging studies of national CI and for centers’ installations of Open XDMoD for studies of local CI.

Original languageEnglish
Article number462
JournalSN Computer Science
Volume5
Issue number5
DOIs
StatePublished - Jun 2024

Keywords

  • Cyberinfrastructure
  • Jupyter notebooks
  • Metrics
  • Open XDMoD

Fingerprint

Dive into the research topics of 'The Data Analytics Framework for XDMoD'. Together they form a unique fingerprint.

Cite this