AndroStep: Android storage performance analysis tool
| dc.contributor.author | Jeong, Sooman | |
| dc.contributor.author | Lee, Kisung | |
| dc.contributor.author | Hwang, Jungwoo | |
| dc.contributor.author | Lee, Seongjin | |
| dc.contributor.author | Won, Youjip | |
| dc.contributor.editor | Wagner, Stefan | |
| dc.contributor.editor | Lichter, Horst | |
| dc.date.accessioned | 2018-10-24T10:00:06Z | |
| dc.date.available | 2018-10-24T10:00:06Z | |
| dc.date.issued | 2013 | |
| dc.description.abstract | The applications in Android based smartphones generate unique IO requests; however, existing IO workload generators and trace capturing tools are designed to neither generate nor capture the IO requests of Android apps. In this paper, we introduce the Android storage performance analysis Tool (AndroStep) which is specifically designed for characterizing and analyzing the behavior of the IO subsystem in Android based devices. AndroStep consists of workload generator, called Mo- bibench, and workload analyzer, called Mobile Storage Analyzer (MOST). Mobibench is an android App, which generates typical filesystem workloads, e.g., Random vs. Sequential and Synchronous vs. Buffered IO, as well as the most dominant workload in Android platform: SQLite insert/update and a write followed by fsync() call. Mobibench can also vary the number of concurrent threads to examine the storage and filesystem overhead to support concurrency, e.g., metadata updates, journal file creation/deletion. MOST capture the trace and extracts key filesystem access characteristics: access pattern with respect to file types, ratio between random vs. sequential access, ratio between buffered vs. synchronous IO, fraction of metadata accesses, etc. MOST implements reverse mapping feature (finding an inode for a given block) and retrospective reverse mapping (finding an inode for a deleted file). We explain the structure and usage of AndroStep in detail. We verify performance result of Mobibench on eight smartphone models. | en |
| dc.identifier.isbn | 978-3-88579-609-1 | |
| dc.identifier.pissn | 1617-5468 | |
| dc.identifier.uri | https://dl.gi.de/handle/20.500.12116/17383 | |
| dc.language.iso | en | |
| dc.publisher | Gesellschaft für Informatik e.V. | |
| dc.relation.ispartof | Software Engineering 2013 - Workshopband | |
| dc.relation.ispartofseries | Lecture Notes in Informatics (LNI) - Proceedings, Volume P-215 | |
| dc.title | AndroStep: Android storage performance analysis tool | en |
| dc.type | Text/Conference Paper | |
| gi.citation.endPage | 340 | |
| gi.citation.publisherPlace | Bonn | |
| gi.citation.startPage | 327 | |
| gi.conference.date | 26. Februar-1. März 2013 | |
| gi.conference.location | Aachen | |
| gi.conference.sessiontitle | Regular Research Papers |
Dateien
Originalbündel
1 - 1 von 1