ACM Transactions on Programming Languages and Systems (TOPLAS)
Sharing memory robustly in message-passing systems
Journal of the ACM (JACM)
Fault-tolerant wait-free shared objects
Journal of the ACM (JACM)
DISC '02 Proceedings of the 16th International Conference on Distributed Computing
Obstruction-Free Synchronization: Double-Ended Queues as an Example
ICDCS '03 Proceedings of the 23rd International Conference on Distributed Computing Systems
Efficient Byzantine-Tolerant Erasure-Coded Storage
DSN '04 Proceedings of the 2004 International Conference on Dependable Systems and Networks
Distributed Computing
Optimal Resilience for Erasure-Coded Byzantine Distributed Storage
DSN '06 Proceedings of the International Conference on Dependable Systems and Networks
Lucky Read/Write Access to Robust Atomic Storage
DSN '06 Proceedings of the International Conference on Dependable Systems and Networks
How fast can a very robust read be?
Proceedings of the twenty-fifth annual ACM symposium on Principles of distributed computing
Tolerating Byzantine Faulty Clients in a Quorum System
ICDCS '06 Proceedings of the 26th IEEE International Conference on Distributed Computing Systems
Wait-free regular storage from Byzantine components
Information Processing Letters
Proceedings of the twenty-sixth annual ACM symposium on Principles of distributed computing
Low-overhead byzantine fault-tolerant storage
Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles
On the Time-Complexity of Robust and Amnesic Storage
OPODIS '08 Proceedings of the 12th International Conference on Principles of Distributed Systems
DISC'07 Proceedings of the 21st international conference on Distributed Computing
DISC'07 Proceedings of the 21st international conference on Distributed Computing
The complexity of robust atomic storage
Proceedings of the 30th annual ACM SIGACT-SIGOPS symposium on Principles of distributed computing
Hi-index | 0.00 |
We present algorithms that reduce the time complexity and improve the scalability of robust storage for unauthenticated data. Robust storage ensures progress under every condition (wait-freedom) and never returns an outdated value (regularity) nor a forged value (Byzantine fault tolerance). The algorithms use secret tokens , which are values randomly selected by the clients and attached to the data written into the storage. Tokens are secret because they cannot be predicted by the attacker before they are used, and thus revealed, by the clients. Our algorithms do not rely on unproven cryptographic assumptions as algorithms based on self-verifying data. They are optimally-resilient, and ensure that reads complete in two communication rounds if readers do not write into the storage, or in one communication round otherwise.