Tag: GPU resource management
Browse our exclusive articles!
ALTO: Fast Adaptive LoRA Tuning for Efficient GPU Use
ALTO accelerates LoRA hyperparameter tuning with adaptive orchestration, boosting GPU efficiency and achieving up to 13.8× speedup without quality loss.
Adaptive Serverless Resource Management to Cut Cold Starts
Optimize serverless computing with adaptive resource management, reducing cold start latency and improving cost-efficiency using predictive and event-drive...
Boost Execution Concurrency in Partial-Order Plans
Enhance AI planning efficiency by improving execution concurrency in partial-order plans using block-substitution and non-concurrency constraints.
Efficient Batch Query Routing for Large Language Models
Optimize large language model queries with batch-level routing under cost, GPU, and capacity constraints for improved performance and resource use.