Plda+ parallel latent dirichlet allocation with data placement and pipeline processing
ACM Transactions on Intelligent Systems and Technology (TIST), 2011•dl.acm.org
Previous methods of distributed Gibbs sampling for LDA run into either memory or
communication bottlenecks. To improve scalability, we propose four strategies: data
placement, pipeline processing, word bundling, and priority-based scheduling. Experiments
show that our strategies significantly reduce the unparallelizable communication bottleneck
and achieve good load balancing, and hence improve scalability of LDA.
communication bottlenecks. To improve scalability, we propose four strategies: data
placement, pipeline processing, word bundling, and priority-based scheduling. Experiments
show that our strategies significantly reduce the unparallelizable communication bottleneck
and achieve good load balancing, and hence improve scalability of LDA.
Previous methods of distributed Gibbs sampling for LDA run into either memory or communication bottlenecks. To improve scalability, we propose four strategies: data placement, pipeline processing, word bundling, and priority-based scheduling. Experiments show that our strategies significantly reduce the unparallelizable communication bottleneck and achieve good load balancing, and hence improve scalability of LDA.
ACM Digital Library