<span class="var-sub_title">Monitoring Parsl Workflows</span> SC18 Proceedings

The International Conference for High Performance Computing, Networking, Storage, and Analysis

Monitoring Parsl Workflows

Student: Connor Pigg (University of Illinois)
Supervisor: Daniel Katz (University of Illinois)

Abstract: As a Python library that enables workflows, Parsl gives users the ability to define complex workflows in Python and run them in parallel on any computer system. This poster describe the process of adding monitoring to Parsl. Simple and comprehensive monitoring of a workflow’s state and resource usage lets users audit, debug, and confirm workflow execution. The poster discusses how Parsl monitors workflow components, what data it captures (task status and resource usage), and the tools it used to do so (Elasticsearch) and to display the information (Kibana). A Kibana dashboard visualizes the collected logs in a real time, with an interactive user interface. This enhanced Parsl allows users the option to monitor the status and resource usage of their workflows via an Elasticsearch database and Kibana dashboard.

ACM-SRC Semi-Finalist: no

Poster: PDF
Poster Summary: pdf

Back to Poster Archive Listing