Thwarting market specific attacks in cloud

Market oriented methodologies have been extensively used for solving dynamic allocation problems in online systems including the Cloud. Despite their extensive use, very little has been known about their security against market specific security threats (e.g. monopoly, shill bidding, etc.). This work follows an experimental driven approach for: (i) promoting the development of threat-aware, market-oriented Clouds, (ii) exposing existing market specific security vulnerabilities and (iii) developing security mechanisms for online markets. We show that the designs of existing market-oriented Clouds are limited when facing market specific attacks and when thwarting malicious bidders and sellers from manipulating auction mechanisms for personal gains. Furthermore, we show that our solutions can effectively resolve market specific attacks and secure bidders, sellers and auctioning mechanisms in the context of Cloud.