4 papers · 1 filter
RapidGNN: Energy and Communication-Efficient Distributed Training on Large-Scale Graph Neural Networks
Arefin Niam, Tevfik Kosar, M S Q Zulkar Nine
Graph Neural Networks (GNNs) have become popular across a diverse set of tasks in exploring structural relationships between entities. However, due to the highly connected structur…
EMLIO: Minimizing I/O Latency and Energy Consumption for Large-Scale AI Training
Hasibul Jamil, MD S Q Zulkar Nine, Tevfik Kosar
Large-scale deep learning workloads increasingly suffer from I/O bottlenecks as datasets grow beyond local storage capacities and GPU compute outpaces network and disk latencies. W…
Carbon-Aware Temporal Data Transfer Scheduling Across Cloud Datacenters
Elvis Rodrigues, Jacob Goldverg, Tevfik Kosar
Inter-datacenter communication is a significant part of cloud operations and produces a substantial amount of carbon emissions for cloud data centers, where the environmental impac…
Energy-Efficient and High-Performance Data Transfers with DRL Agents
Hasibul Jamil, Jacob Goldverg, Elvis Rodrigues +2
The rapid growth of data across fields of science and industry has increased the need to improve the performance of end-to-end data transfers while using the resources more efficie…