Experience with Grapevine: the growth of a distributed system
ACM Transactions on Computer Systems (TOCS)
The process group approach to reliable distributed computing
Communications of the ACM
Weak-consistency group communication and membership
Weak-consistency group communication and membership
DEC data distributor: for data replication and data warehousing
SIGMOD '94 Proceedings of the 1994 ACM SIGMOD international conference on Management of data
Flexible update propagation for weakly consistent replication
Proceedings of the sixteenth ACM symposium on Operating systems principles
Computer rendering of stochastic models
Communications of the ACM
On the geographic location of internet resources
Proceedings of the 2nd ACM SIGCOMM Workshop on Internet measurment
A Demand based Algorithm for Rapid Updating of Replicas
ICDCSW '02 Proceedings of the 22nd International Conference on Distributed Computing Systems
Weak consistency: a generalized theory and optimistic implementations for distributed transactions
Weak consistency: a generalized theory and optimistic implementations for distributed transactions
Generalization of the fast consistency algorithm to a grid with multiple high demand zones
ICCS'03 Proceedings of the 2003 international conference on Computational science: PartII
| Hi-index | 0.00 |
In this paper we evaluate our own weak consistency algorithm, which is called the ”Fast Consistency Algorithm”, and whose main aim is optimizing the propagation of changes introducing a preference for nodes and zones of the network which have greatest demand. Weak consistency algorithms allow us to propagate changes in a large, arbitrary changing storage network in a self-organizing way. These algorithms generate very little traffic overhead; they have low latency and are scalable, in addition to being fault tolerant. The algorithm has been simulated over ns-2, and measured its performance for complex spatial distributions of demand, including Internet like self-similar fractal distributions of demand. The impulse response of the algorithm has been characterized. We conclude that considering application parameters such as demand in the event or change propagation mechanism to: 1) prioritize probabilistic interactions with neighbors with higher demand, and 2) including little changes on the logical topology (leader interconnection in hierarchical topology ), gives a surprising improvement in the speed of change propagation perceived by most users. In other words, it satisfies the greatest demand in the shortest amount of time.