TY - GEN
T1 - Ebserver
T2 - 10th IEEE/ACM International Conference on Mobile Software Engineering and Systems, MOBILESoft 2023
AU - Oliveira, Wellington
AU - Moraes, Bernardo
AU - Castor, Fernando
AU - Fernandes, Joao Paulo
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Mobile applications are a typical component of people's routines. Because of that, there is fierce competition for mobile users' attention, creating pressure for mobile developers to optimize their applications in a number of ways, such as making them faster, reducing their energy consumption, or their memory usage. To understand their application resource usage, developers need to execute their app, collect data from that execution and analyze how it behaves. Researchers must also go through this process when evaluating optimizations and techniques to reduce resource usage. This error-prone experimentation process can take hours of repetitive work if done manually. In this paper, we present EBSERVER, a general-purpose measurement automation tool to collect Android device data during application executions. EBSERVER is simple to configure and extend, requiring very little instrumentation code to use. It enables users to collect execution metrics on a per-process basis from an application execution automatically. Examples of such metrics include energy consumption, CPU usage, execution time, and memory usage. EBSERVER makes it possible for applications to run multiple times in an automated manner, eliminates the need to predict the time that applications or benchmarks will run in an experiment, and is compatible with contemporary Android UI testing tools. EBSERVER has been employed in multiple experiments, including experiments that do not have involvement of its authors.
AB - Mobile applications are a typical component of people's routines. Because of that, there is fierce competition for mobile users' attention, creating pressure for mobile developers to optimize their applications in a number of ways, such as making them faster, reducing their energy consumption, or their memory usage. To understand their application resource usage, developers need to execute their app, collect data from that execution and analyze how it behaves. Researchers must also go through this process when evaluating optimizations and techniques to reduce resource usage. This error-prone experimentation process can take hours of repetitive work if done manually. In this paper, we present EBSERVER, a general-purpose measurement automation tool to collect Android device data during application executions. EBSERVER is simple to configure and extend, requiring very little instrumentation code to use. It enables users to collect execution metrics on a per-process basis from an application execution automatically. Examples of such metrics include energy consumption, CPU usage, execution time, and memory usage. EBSERVER makes it possible for applications to run multiple times in an automated manner, eliminates the need to predict the time that applications or benchmarks will run in an experiment, and is compatible with contemporary Android UI testing tools. EBSERVER has been employed in multiple experiments, including experiments that do not have involvement of its authors.
KW - Android
KW - Energy
KW - Metrics
KW - Performance
KW - Tool
UR - http://www.scopus.com/inward/record.url?scp=85162189107&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85162189107&partnerID=8YFLogxK
U2 - 10.1109/MOBILSoft59058.2023.00014
DO - 10.1109/MOBILSoft59058.2023.00014
M3 - Conference contribution
AN - SCOPUS:85162189107
T3 - Proceedings - 2023 IEEE/ACM 10th International Conference on Mobile Software Engineering and Systems, MOBILESoft 2023
SP - 55
EP - 59
BT - Proceedings - 2023 IEEE/ACM 10th International Conference on Mobile Software Engineering and Systems, MOBILESoft 2023
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 14 May 2023 through 15 May 2023
ER -