Performance Antipatterns: Angel or Devil for Power Consumption?
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918334583275520 |
|---|---|
| author | Aneggi, Alessandro Stoico, Vincenzo Janes, Andrea |
| author_facet | Aneggi, Alessandro Stoico, Vincenzo Janes, Andrea |
| contents | Performance antipatterns are known to degrade the responsiveness of microservice-based systems, but their impact on energy consumption remains largely unexplored. This paper empirically investigates whether widely studied performance antipatterns defined by Smith and Williams also negatively influence power usage. We implement ten antipatterns as isolated microservices and evaluate them under controlled load conditions, collecting synchronized measurements of performance, CPU and DRAM power consumption, and resource utilization across 30 repeated runs per antipattern. The results show that while all antipatterns degrade performance as expected, only a subset exhibit a statistically significant relationship between response time and increased power consumption. Specifically, several antipatterns reach CPU saturation, capping power draw regardless of rising response time, whereas others (\eg Unnecessary Processing, The Ramp) demonstrate energy-performance coupling indicative of inefficiency. Our results show that, while all injected performance antipatterns increase response time as expected, only a subset also behaves as clear energy antipatterns, with several cases reaching a nearly constant CPU power level where additional slowdowns mainly translate into longer execution time rather than higher instantaneous power consumption. The study provides a systematic foundation for identifying performance antipatterns that also behave as energy antipatterns and offers actionable insights for designing more energy-efficient microservices architectures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_12079 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Performance Antipatterns: Angel or Devil for Power Consumption? Aneggi, Alessandro Stoico, Vincenzo Janes, Andrea Software Engineering Performance antipatterns are known to degrade the responsiveness of microservice-based systems, but their impact on energy consumption remains largely unexplored. This paper empirically investigates whether widely studied performance antipatterns defined by Smith and Williams also negatively influence power usage. We implement ten antipatterns as isolated microservices and evaluate them under controlled load conditions, collecting synchronized measurements of performance, CPU and DRAM power consumption, and resource utilization across 30 repeated runs per antipattern. The results show that while all antipatterns degrade performance as expected, only a subset exhibit a statistically significant relationship between response time and increased power consumption. Specifically, several antipatterns reach CPU saturation, capping power draw regardless of rising response time, whereas others (\eg Unnecessary Processing, The Ramp) demonstrate energy-performance coupling indicative of inefficiency. Our results show that, while all injected performance antipatterns increase response time as expected, only a subset also behaves as clear energy antipatterns, with several cases reaching a nearly constant CPU power level where additional slowdowns mainly translate into longer execution time rather than higher instantaneous power consumption. The study provides a systematic foundation for identifying performance antipatterns that also behave as energy antipatterns and offers actionable insights for designing more energy-efficient microservices architectures. |
| title | Performance Antipatterns: Angel or Devil for Power Consumption? |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2602.12079 |