Current computing systems employ different mechanisms to deal with overload conditions. Of those widely deployed are content adaptation mechanisms whereby the quality level of the content is adapted dynamically to mitigate overload conditions. Serving degraded content reduces strain on resources and enables them to cater for a larger set of clients. To that end, this paper studies adversarial exploits of dynamic content adaptation mechanisms to a new instantiation of Reduction of Quality (RoQ) attacks. The RoQ attack pattern is orchestrated to cause different forms of damage such as longer response time for legitimate clients, degraded content being served and under-utilization of resources. We assess the impact of RoQ attacks via the potency metric which reflects the tradeoffs between the damage inflicted and the cost in mounting the attack. We validate our results through numerical analysis as well as real Internet experiments.
展开▼
