A Dashboard for Automatic Monitoring Python Web Services

Vogel, P (2017) A Dashboard for Automatic Monitoring Python Web Services. Bachelor's Thesis, Computing Science.

Preview	Text bsc-thesis3.pdf - Published Version Download (997kB) | Preview
	Text Toestemming.pdf - Other Restricted to Backend only Download (80kB)

Abstract

This bachelor thesis describes the problem of monitoring the performance of Flask-based Python web-services. For the web developer that wants to monitor the performance for their web-services, a solution is presented. The solution consists of an automatic monitoring dashboard, that can be installed in any existing web-service with Python and Flask. After the installation (installation requires 2 lines of code) an automatic monitoring service is ready to use. With several lines of extra configuration, the following features are supported: - Automatic version detection. The monitoring tool detects the active version of this VCS and combines this with the collected data. - Comparison of execution times across different users. Which users perform better or worse on certain versions of the system and how can this be improved? The monitoring dashboard creates graphs automatically, wherein it is easy to spot differences in execution times. - Automatic outlier detection. Whenever the execution time is larger than usual -- an outlier --, the monitoring tool collects extra information about the requested data, such as the stack-trace of active threads, and CPU- and memory usage of the system. In order to reduce the overhead of the dashboard, logging extra information is only done for potential outliers. Moreover, this thesis describes the design of the monitoring tool. After the development of the dashboard, it is deployed for a case study to validate its usefulness. The collected results have been used to analyze and improve the performance of that case study.

Item Type:	Thesis (Bachelor's Thesis)
Degree programme:	Computing Science
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	15 Feb 2018 08:30
Last Modified:	15 Feb 2018 08:30
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/15605

Actions (login required)

View Item