Queue Prediction
Cross-source consensus on Queue Prediction from 1 sources and 5 claims.
1 sources · 5 claims
How it works
Benefits
Comparisons
Evidence quality
Highlighted claims
- FedQueue estimates queue delays with an exponentially weighted moving average updated when client updates arrive. — FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training
- Online queue estimates allow FedQueue to reduce local work during predicted long queues and increase it when admission is expected to be faster. — FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training
- Removing EWMA greatly increased time to target accuracy and worsened normalized expected delay in ablation experiments. — FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training
- FedQueue's theory is not tied to EWMA specifically and requires sub-Gaussian queue-prediction errors. — FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training
- Moderate queue-predictor smoothing performed better than overly slow or overly reactive estimates in the reported experiments. — FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training