Deeply-Pipelined FPGA Clusters Make DNN Training Scalable

Deep Neural Networks (DNNs) have revolutionized numerous applications, but the demand for ever more performance remains unabated. Scaling DNN computations to larger clusters is generally done by distributing tasks in batch mode using methods such as distributed synchronous SGD. Among the issues with this approach is that to make the distributed cluster work with high utilization, the workload distributed to each node must be large, which implies nontrivial growth in the SGD mini-batch size. In this talk, we introduce a framework, FPDeep, which uses a hybrid of model and layer parallelism to configure distributed reconfigurable clusters to train DNNs. This approach has numerous benefits. First, the design does not suffer from batch size growth. Second, novel workload and weight partitioning leads to balanced loads of both among nodes. And third, the entire system is a fine-grained pipeline. This leads to high parallelism and utilization and also minimizes the time features need to be cached while waiting for back-propagation. As a result, storage demand is reduced to the point where only on-chip memory is used for the convolution layers. We evaluate FPDeep with the Alexnet, VGG-16, and VGG-19 benchmarks. FPDeep provides, on average, 6.36x higher energy efficiency than comparable GPU servers.

Event Name

IXPUG Webinar Series


Data Parallelism,Inference,FPGA,FPDeep,Convolutional Neural Networks (CNN),Convolutional Neural Networks (CNN) Training,Hybrid Model/Layer Parallelism,Workload Partitioning