{"title":"Modeling Availability in Clouds for Mobile Computing","authors":"Suraj Pandey, S. Nepal","doi":"10.1109/MobServ.2012.11","DOIUrl":null,"url":null,"abstract":"CIA - confidentiality, integrity, and availability of data are the three most desirable characteristics of a data Cloud. Availability, in particular, is interesting when it comes to hosting data of endusers who are always on the move (mobile clients).Conservatively, a full replication of data services would be ideal. However, in a competitive market, where cost and quality of services play a major role in determining the existence of IT services, a total replication-based solution will not yield the best service nor will it be economically viable to the endusers. Especially, for those users who are mobile and would want their data available anywhere around the world at any time, the optimization of data service availability and cost of keeping it in the range 90%to 99.9% is challenging. This paper defines an availability metric and proposes an optimization model that makes use of multiple Cloud data services for making data services highly available at lower cost. By using user-defined availability profiles, the revenue expected by a cloud data service provider is shaped accordingly. When the availability metric, user-defined availability profile, the cost model, and the data service utilization are put together, mobile end-users get the best out of their data services and cost of making them available anywhere anytime using the Cloud deployment model.","PeriodicalId":443858,"journal":{"name":"2012 IEEE First International Conference on Mobile Services","volume":"22 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 IEEE First International Conference on Mobile Services","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MobServ.2012.11","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 14
Abstract
CIA - confidentiality, integrity, and availability of data are the three most desirable characteristics of a data Cloud. Availability, in particular, is interesting when it comes to hosting data of endusers who are always on the move (mobile clients).Conservatively, a full replication of data services would be ideal. However, in a competitive market, where cost and quality of services play a major role in determining the existence of IT services, a total replication-based solution will not yield the best service nor will it be economically viable to the endusers. Especially, for those users who are mobile and would want their data available anywhere around the world at any time, the optimization of data service availability and cost of keeping it in the range 90%to 99.9% is challenging. This paper defines an availability metric and proposes an optimization model that makes use of multiple Cloud data services for making data services highly available at lower cost. By using user-defined availability profiles, the revenue expected by a cloud data service provider is shaped accordingly. When the availability metric, user-defined availability profile, the cost model, and the data service utilization are put together, mobile end-users get the best out of their data services and cost of making them available anywhere anytime using the Cloud deployment model.