DEFT: Joint Task Placement and DVFS for Energy-Efficient Multi-GPU Runtimes
2026-08-03 • Distributed, Parallel, and Cluster Computing
Distributed, Parallel, and Cluster Computing
AI summaryⓘ
The authors address the problem of saving energy in computers that use multiple GPUs to speed up tasks. They created DEFT, a system that smartly decides where to run tasks and how fast each GPU should run, all while considering how tasks move between GPUs and the cost of changing speeds. Unlike older methods, DEFT looks at these factors together at a very detailed level to make better energy and performance decisions. Tests showed that DEFT reduces energy use and improves efficiency without slowing things down much.
Energy EfficiencyGPUDynamic Voltage and Frequency Scaling (DVFS)Task SchedulingMulti-GPU SystemsTask PlacementInter-GPU CommunicationCUDAEnergy-Delay Product (EDP)Runtime Systems
Authors
Jing Chen, Miquel Pericas
Abstract
Energy efficiency has become a first-order concern in modern high-performance computing systems, as it directly determines achievable throughput under fixed power budgets. Although Dynamic Voltage and Frequency Scaling (DVFS) provides an effective mechanism for reducing GPU energy consumption, existing runtime systems decouple DVFS from task placement and inter-GPU communication, focus on single-GPU execution, or cannot adapt frequency to task granularity and runtime contention in multi-GPU environments. Consequently, current schedulers fail to capture the tight coupling between task placement, frequency selection, and inter-GPU data movement that fundamentally governs energy-performance trade-offs on multi-GPU systems. This paper presents DEFT, an energy-aware scheduling framework that jointly optimizes task-to-device assignment and per-GPU DVFS configuration for task-based multi-GPU applications. DEFT employs a cost-model-driven strategy that integrates slack awareness, throughput awareness, and explicit modeling of task execution cost, inter-GPU data movement, and DVFS transition overheads, enabling coordinated placement and frequency decisions at task granularity under dynamic runtime conditions. We prototype DEFT within the CUDASTF runtime and demonstrate its effectiveness across five optimization objectives. The evaluation shows that DEFT reduces energy consumption by 14.8% and 4.8% on average on NVIDIA L40S and L4, and reduces EDP by 9.9% and 3.7%, respectively, while maintaining performance within 1.5% of the fastest baseline.