{"title":"Evolution of the BFQ Storage-I/O Scheduler","authors":"P. Valente, Arianna Avanzini","doi":"10.1109/MST.2015.9","DOIUrl":null,"url":null,"abstract":"An accurate storage-I/O scheduler, named Budget Fair Queueing (BFQ), was integrated with a special set of heuristics a few years ago. The resulting, improved scheduler, codenamed BFQ-v1, was able to guarantee a number of desirable service properties, including a high responsiveness, to applications and system services. In the intervening years, BFQ-v1 has become relatively popular on desktop and handheld systems, and has further evolved. But no official, comprehensive and concentrated documentation has been provided about the improvements that have followed each other. In this paper we fill this documentation gap, by describing the current, last version of BFQ (v7r8). We also show the performance of BFQ-v7r8 through some experimental results, in terms of throughput and application responsiveness, and on both an HDD and an SSD.","PeriodicalId":252572,"journal":{"name":"2015 Mobile Systems Technologies Workshop (MST)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 Mobile Systems Technologies Workshop (MST)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MST.2015.9","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
An accurate storage-I/O scheduler, named Budget Fair Queueing (BFQ), was integrated with a special set of heuristics a few years ago. The resulting, improved scheduler, codenamed BFQ-v1, was able to guarantee a number of desirable service properties, including a high responsiveness, to applications and system services. In the intervening years, BFQ-v1 has become relatively popular on desktop and handheld systems, and has further evolved. But no official, comprehensive and concentrated documentation has been provided about the improvements that have followed each other. In this paper we fill this documentation gap, by describing the current, last version of BFQ (v7r8). We also show the performance of BFQ-v7r8 through some experimental results, in terms of throughput and application responsiveness, and on both an HDD and an SSD.