{"title":"Visualizing Multi-process CPU Utilization using CUSP","authors":"M. Maimone","doi":"10.1109/AERO53065.2022.9843349","DOIUrl":null,"url":null,"abstract":"The CPU Utilization Statistics Plotter (CUSP) tool automates the interpretation of detailed CPU Utilization trace data and statistics. It puts you on the cusp of understanding how CPU resources are split among the many parallel components of a software system. CUSP combines time-sampled CPU utilization numbers and Event Log annotations to generate human-readable plots and tables. It automatically splits up large CPU usage log files around interesting events, determines and highlights just the tasks of primary relevance by evaluating their changing contribution to each plot's total CPU usage, automatically eliminates irrelevant tasks, provides context by labeling plots with names and durations of all active commands, and uses consistent color-coding to enable quick visual comparison across multiple plots. CUSP has been used to process CPU Utilization trace logs on the Mars Science Laboratory and the Mars 2020 Rover missions during flight software development and Flight Operations on the Martian surface since December 2013.","PeriodicalId":219988,"journal":{"name":"2022 IEEE Aerospace Conference (AERO)","volume":"407 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE Aerospace Conference (AERO)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AERO53065.2022.9843349","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
The CPU Utilization Statistics Plotter (CUSP) tool automates the interpretation of detailed CPU Utilization trace data and statistics. It puts you on the cusp of understanding how CPU resources are split among the many parallel components of a software system. CUSP combines time-sampled CPU utilization numbers and Event Log annotations to generate human-readable plots and tables. It automatically splits up large CPU usage log files around interesting events, determines and highlights just the tasks of primary relevance by evaluating their changing contribution to each plot's total CPU usage, automatically eliminates irrelevant tasks, provides context by labeling plots with names and durations of all active commands, and uses consistent color-coding to enable quick visual comparison across multiple plots. CUSP has been used to process CPU Utilization trace logs on the Mars Science Laboratory and the Mars 2020 Rover missions during flight software development and Flight Operations on the Martian surface since December 2013.