Authors:

Nayeli Gurrola

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

Effective collecting, management, and analysis of performance data on Purdue University's Anvil supercomputer is crucial for system administrators to gain valuable insights into software usage and performance metrics. XALT is an instrumental tool in capturing data related to software dependencies and execution on the Anvil system. In this project, we leveraged XALT alongside Python and web-based visualization frameworks to develop an interactive dashboard. The dashboard empowers system administrators to explore and analyze the Anvil software ecosystem, facilitating data-driven decision-making for optimizing system resources and enhancing user experiences. We extracted and transformed relevant data from XALT using SQL and Python, revealing valuable insights. These insights were presented interactively using Dash, a Python framework for building interactive web applications. By integrating XALT, SQL, Python, and Dash, our project enables system administrators to comprehensively understand software usage patterns and system utilization of the Anvil supercomputer. Our project showcases the potential of XALT and modern technologies in analyzing and visualizing performance data. It contributes to system administration, system optimization as well as better user support in high-performance computing (HPC) environments, specifically on Purdue University's Anvil supercomputer.

Source:

Purdue University / 2023

Topics:

No topics listed

Co-authors:

Nayeli Gurrola