2 papers
cs.AR2019
AMOEBA: A Coarse Grained Reconfigurable Architecture for Dynamic GPU Scaling
Xianwei Cheng, Hui Zhao, Mahmut Kandemir +2
Different GPU applications exhibit varying scalability patterns with network-on-chip (NoC), coalescing, memory and control divergence, and L1 cache behavior. A GPU consists of seve…
cs.DC2019
Alleviating Bottlenecks for DNN Execution on GPUs via Opportunistic Computing
Xianwei Cheng, Hui Zhao, Mahmut Kandemir +2
Edge computing and IoT applications are severely constrained by limited hardware resource. This makes memory consuming DNN frameworks not applicable to edge computing. Simple algor…