Decoy routing is a powerful circumvention mechanism intended to provide secure communications that cannot be monitored, detected, or disrupted by a third party who controls the user's network infrastructure. Current decoy routing protocols have weaknesses, however: they either make the unrealistic assumption that routes through the network are symmetric (i.e., the router implementing the decoy routing protocol must see all of the traffic, in both directions, from each connection it modifies), or their protocol requires modifying the route taken by packets in connections that use the protocol, and these route changes are detectable by a third party. We present Rebound, a decoy routing protocol that tolerates asymmetric routes without modifying the route taken by any packet that passes through the decoy router, making it more difficult to detect or disrupt than previous decoy routing protocols.

No abstract available.

DISP is a practical client-server protocol for the distributed storage of immutable data objects. Unlike most other contemporary protocols, DISP permits applications to make explicit tradeoffs between total storage space, computational overhead, and guarantees of availability, integrity, and privacy on a per-object basis. Applications specify the degree of redundancy with which each item is encoded, what level of integrity checks are computed and stored with each item, and whether items are stored in an encrypted format. At one extreme, clients willing to pay the overhead are guaranteed privacy, integrity, and availability of data stored in the system as long as fewer than half the servers are Byzantine. At the other extreme, objects that do not require privacy or integrity in the face of Byzantine servers can be stored with very low computational and storage overhead.DISP is efficient in terms of message count, message size, and storage requirements: even in the worst case, the read and write protocols require a number of messages that are linear with respect to the number of servers. In terms of message size, DISP requires transferring only marginally more than L bytes to correctly read an object of size L , even in the face of Byzantine server failures. In this article we provide a description of DISP and an analysis of its fault-tolerant properties. We also analyze the complexity of the protocol and discuss several potential applications. We conclude with a description of our prototype implementation and measurements of its performance on commodity hardware.