Abstract: A method for reducing latency and increasing throughput in a reconfigurable computing system includes receiving a compute graph for execution on a reconfigurable dataflow processor comprising a grid of compute units and grid of memory units interconnected with a switching array. The compute graph includes a node specifying an operation on a tensor. The node may be split into multiple nodes that each specify the operation on a distinctive portion of the tensor to produce a first modified compute graph. The first modified compute graph may be executed. In addition, the multiple nodes may be within a single meta-pipeline stage and may be processed in parallel. Furthermore, the compute graph may further comprise a separate node for gathering the distinctive portions of the tensor into a complete tensor, to produce a second modified compute graph.

Abstract: A videoconference system and a management method thereof are provided. The videoconference system includes: terminals; multipoint control units (MCUs), as the superior level for the terminals, for controlling the corresponding terminals; conference management systems, as the superior level for the MCUs, for performing dispatching and control for the local conference site; and conference coordination systems, as the superior level for the conference management systems, for coordinating corresponding conference management systems and neighboring conference coordination systems to perform dispatching and control for the entire conference.