{"title":"基于跟踪基准的Android系统可重复负载测试","authors":"Alexander Lochmann, Fabian Bruckner, O. Spinczyk","doi":"10.1145/3053600.3053615","DOIUrl":null,"url":null,"abstract":"The development of system software and hardware components for Android devices is strongly influenced by the necessity to save energy. However, there is no methodology that provides developers with reproducible and comparable benchmarks for testing the device under a representative load. Such a benchmark would have to stimulate all relevant parts of the system and must neither depend on the current state of external servers in the Internet nor on any interactive user. This paper describes the first steps towards such a benchmark. The approach is based on recorded workload traces of prominent Android applications. From these traces we can \"mix a cocktail\" that yields a representative workload profile. By replaying the recorded and mixed loads with a workload generator combined with an external environment for dealing with communication workloads we obtain benchmarks that fulfill the requirements.","PeriodicalId":115833,"journal":{"name":"Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering Companion","volume":"215 ","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Reproducible Load Tests for Android Systems with Trace-based Benchmarks\",\"authors\":\"Alexander Lochmann, Fabian Bruckner, O. Spinczyk\",\"doi\":\"10.1145/3053600.3053615\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The development of system software and hardware components for Android devices is strongly influenced by the necessity to save energy. However, there is no methodology that provides developers with reproducible and comparable benchmarks for testing the device under a representative load. Such a benchmark would have to stimulate all relevant parts of the system and must neither depend on the current state of external servers in the Internet nor on any interactive user. This paper describes the first steps towards such a benchmark. The approach is based on recorded workload traces of prominent Android applications. From these traces we can \\\"mix a cocktail\\\" that yields a representative workload profile. By replaying the recorded and mixed loads with a workload generator combined with an external environment for dealing with communication workloads we obtain benchmarks that fulfill the requirements.\",\"PeriodicalId\":115833,\"journal\":{\"name\":\"Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering Companion\",\"volume\":\"215 \",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-04-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering Companion\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3053600.3053615\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 8th ACM/SPEC on International Conference on Performance Engineering Companion","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3053600.3053615","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Reproducible Load Tests for Android Systems with Trace-based Benchmarks
The development of system software and hardware components for Android devices is strongly influenced by the necessity to save energy. However, there is no methodology that provides developers with reproducible and comparable benchmarks for testing the device under a representative load. Such a benchmark would have to stimulate all relevant parts of the system and must neither depend on the current state of external servers in the Internet nor on any interactive user. This paper describes the first steps towards such a benchmark. The approach is based on recorded workload traces of prominent Android applications. From these traces we can "mix a cocktail" that yields a representative workload profile. By replaying the recorded and mixed loads with a workload generator combined with an external environment for dealing with communication workloads we obtain benchmarks that fulfill the requirements.