Linearizability: a correctness condition for concurrent objects
ACM Transactions on Programming Languages and Systems (TOPLAS)
Renaming in an asynchronous environment
Journal of the ACM (JACM)
ACM Transactions on Programming Languages and Systems (TOPLAS)
Atomic snapshots of shared memory
Journal of the ACM (JACM)
Immediate atomic snapshots and fast renaming
PODC '93 Proceedings of the twelfth annual ACM symposium on Principles of distributed computing
Generalized FLP impossibility result for t-resilient asynchronous computations
STOC '93 Proceedings of the twenty-fifth annual ACM symposium on Theory of computing
Atomic Snapshots in O (n log n) Operations
SIAM Journal on Computing
The topological structure of asynchronous computability
Journal of the ACM (JACM)
Wait-Free k-Set Agreement is Impossible: The Topology of Public Knowledge
SIAM Journal on Computing
Obstruction-Free Synchronization: Double-Ended Queues as an Example
ICDCS '03 Proceedings of the 23rd International Conference on Distributed Computing Systems
Distributed Computing: Fundamentals, Simulations and Advanced Topics
Distributed Computing: Fundamentals, Simulations and Advanced Topics
Test & Set, Adaptive Renaming and Set Agreement: a Guided Visit to Asynchronous Computability
SRDS '07 Proceedings of the 26th IEEE International Symposium on Reliable Distributed Systems
Proceedings of the twentieth annual symposium on Parallelism in algorithms and architectures
New combinatorial topology upper and lower bounds for renaming
Proceedings of the twenty-seventh ACM symposium on Principles of distributed computing
Locks Considered Harmful: A Look at Non-traditional Synchronization
SEUS '08 Proceedings of the 6th IFIP WG 10.2 international workshop on Software Technologies for Embedded and Ubiquitous Systems
From adaptive renaming to set agreement
Theoretical Computer Science
Help when needed, but no more: efficient read/write partial snapshot
DISC'09 Proceedings of the 23rd international conference on Distributed computing
Contention-sensitive data structures and algorithms
DISC'09 Proceedings of the 23rd international conference on Distributed computing
In search of the holy grail: looking for the weakest failure detector for wait-free set agreement
OPODIS'06 Proceedings of the 10th international conference on Principles of Distributed Systems
Renaming with k-set-consensus: an optimal algorithm into n + k - 1 slots
OPODIS'06 Proceedings of the 10th international conference on Principles of Distributed Systems
DISC'06 Proceedings of the 20th international conference on Distributed Computing
The renaming problem in shared memory systems: An introduction
Computer Science Review
| Hi-index | 0.00 |
The adaptive M-renaming problem consists in designing an algorithm that allows a set of p ≤ n participating asynchronous processes (where n is the total number of processes) not known in advance to acquire pair-wise different new names in a name space whose size M depends on p (and not on n). Adaptive (2p - 1)-renaming algorithms for read/write shared memory systems have been designed. These algorithms, which are optimal with respect to the value of M, consider the wait-freedom progress condition, which means that any correct participant has to acquire a new name whatever the behavior of the other processes (that can be very slow or even crashed). This paper addresses the design of an adaptive M-renaming algorithm when considering the k-obstruction-freedom progress condition. This condition, that is weaker than wait-freedom, requires that every correct participating process acquires a new name in all runs where during "long enough periods" at most k processes execute steps (p-obstruction-freedom and wait-freedom are actually equivalent). The paper presents an optimal adaptive (p + k - 1)-renaming algorithm and, consequently, contributes to a better understanding of synchronization and concurrency by showing that weakening the liveness condition from wait-freedom to k-obstruction-freedom allows the new name space to be reduced from 2p - 1 to min(2p - 1, p + k - 1). Last but not least, the proposed algorithm is particularly simple, a first class property. This establishes an interesting tradeoff linking progress conditions with the size of the new name space.