Distributed Parallel Inference on Large Factor Graphs
This addresses the need for efficient parallel inference in AI as problem sizes grow, though it appears incremental as it builds on existing methods for distributed settings.
The paper tackled the problem of parallel inference on large factor graphs in distributed memory clusters by developing the DBRSplash algorithm, which achieved linear to super-linear performance gains on a 120-processor cluster.
As computer clusters become more common and the size of the problems encountered in the field of AI grows, there is an increasing demand for efficient parallel inference algorithms. We consider the problem of parallel inference on large factor graphs in the distributed memory setting of computer clusters. We develop a new efficient parallel inference algorithm, DBRSplash, which incorporates over-segmented graph partitioning, belief residual scheduling, and uniform work Splash operations. We empirically evaluate the DBRSplash algorithm on a 120 processor cluster and demonstrate linear to super-linear performance gains on large factor graph models.