A bridging model for parallel computation
Communications of the ACM
Communication-efficient parallel sorting (preliminary version)
STOC '96 Proceedings of the twenty-eighth annual ACM symposium on Theory of computing
LogP: a practical model of parallel computation
Communications of the ACM
Emulations between QSM, BSP, and LogP: a framework for general-purpose parallel algorithm design
Proceedings of the tenth annual ACM-SIAM symposium on Discrete algorithms
On the Effectiveness of D-BSP as a Bridging Model of Parallel Computation
ICCS '01 Proceedings of the International Conference on Computational Science-Part II
A Parallel Computational Model for Heterogeneous Clusters
IEEE Transactions on Parallel and Distributed Systems
Hi-index | 0.00 |
We investigate the issue of stalling in the LogP model. In particular, we introduce a novel quantitative characterization of stalling, referred to as 驴-stalling, which intuitively captures the realistic assumption that once the network's capacity constraint is violated, it takes some time (at most 驴) for this information to propagate to the processors involved.We prove a lower bound that shows that LogP under 驴-stalling is strictly more powerful than the stall-free version of the model where only strictly stall-free computations are permitted. On the other hand, we show that 驴-stalling LogP with 驴 = L can be simulated with at most logarithmic slowdown by a BSP machine with similar bandwidth and latency values, thus extending the equivalence (up to logarithmic factors) between stall-free LogP and BSP argued in [1] to the more powerful L-stalling LogP.