This paper presents the design, implementation, analysis, and experimental evaluation of speak-up, a defense against applicationlevel distributed denial-of-service (DDoS), in which attackers cripple a server by sending legitimate-looking requests that consume computational resources (e.g., CPU cycles, disk). With speak-up, a victimized server encourages all clients, resources permitting, to automatically send higher volumes of traffic. We suppose that attackers are already using most of their upload bandwidth so cannot react to the encouragement. Good clients, however, have spare upload bandwidth and will react to the encouragement with drastically higher volumes of traffic. The intended outcome of this traffic inflation is that the good clients crowd out the bad ones, thereby capturing a much larger fraction of the server’s resources than before. We experiment under various conditions and find that speak-up causes the server to spend resources on a group of clients in rough proportion to their aggregate upload bandwidth. This result makes the defense viable and effective for a class of real attacks.

We present a solution to the denial of service (DoS) problem that does not rely on network infrastructure support, conforming to the end-to-end (e2e) design principle. Our approach is to combine an overlay network, which allows us to treat authorized traffic preferentially, with a lightweight process-migration environment that allows us to move services easily between different parts of a distributed system. Functionality residing on a part of the system that is subjected to a DoS attack migrates to an unaffected location. The overlay network ensures that traffic from legitimate users, who are authenticated before they are allowed to access the service, is routed to the new location. We demonstrate the feasibility and effectiveness of our approach by measuring the performance of an experimental prototype against a series of attacks using PlanetLab, a distributed experimental testbed. Our preliminary results show that the end-toend latency remains at acceptable levels during regular operation, increasing only by a factor of 2 to 3, even for large overlays. When a process migrates due to a DoS attack, the disruption of service for the end user is in the order of a few seconds, depending on the network proximity of the servers involved in the migration.

Distributed Denial Of Service (DDoS) attacks are familiar threats to Internet users for more than ten years. Such attacks are carried out by a “Bot net”, an army of zombie hosts spread around the Internet, that overwhelm the bandwidth toward their victim Web server, by sending traffic upon command. This paper introduces WDA, a novel architecture to attenuate the DDoS attacker’s bandwidth. WDA is especially designed to protect Web farms. WDA is asymmetric and only monitors and protects the uplink toward the Web farm, which is the typical bottleneck in DDoS attacks. Legitimate traffic toward Web farms is very distinctive since it is produced by humans using Web browsing software. Specifically, such upload traffic has low volume, and more importantly, has long off times that correspond to human view time. WDA utilizes these properties of legitimate client traffic to distinguish it from attack traffic, which tends to be continuous and heavy. A key feature of WDA is in its use of randomized thresholds that trap and penalize deterministic zombie traffic that tries to mimic human client patterns. WDA’s heart is WDAQ, a novel active queue management mechanism aimed to prefer legitimate client traffic over attacker traffic. With WDA installed, the attacker traffic toward the victim is attenuated. Extensive simulation results show that WDA can defeat simple flooding attacks, and can attenuate the bandwidth usable by sophisticated WDA-aware attacks by orders of magnitude. As a consequence, the attacker must increase his “bot-net ” size by the same factor, to compensate for the effects of WDA. Our simulations show that WDA can defend a typical Web farm from DDoS attacks launched by hundreds of thousands zombies, while keeping legitimate clients ’ service degradation under ten percent.