計算タスクの完了時間を考慮した計算資源と電力の協調運用

With the rapid growth of computing demand and the large-scale integration of renewable energy, how to realize the coordinated optimal operation of computing networks and power networks has become an important issue to be addressed. However, existing studies mainly focus on the impacts of spatiotemporal migration of computing loads on power system operation, while the internal task processing procedures of computing networks are largely neglected. To address this issue, this paper proposes an optimal operation method for computing power-electric power coordination considering the end-to-end completion latency of computing tasks. First, a full-process latency model covering the "transmission-buffering-computation" procedure of computing tasks is established, which uniformly characterizes forwarding waiting, network transmission, queueing, and computation processing. Second, a task-level spatiotemporal scheduling mechanism is developed to jointly optimize the task forwarding time, routing path, and destination computing node. Then, a computing power-electric power coordinated optimization model is formulated to minimize the power supply cost of the power network and the total completion latency of computing tasks. Case studies demonstrate that the proposed method can fully exploit the task-level spatiotemporal scheduling flexibility of computing loads and facilitate the spatiotemporal matching between computing loads and renewable energy, thereby reducing the system power supply cost while ensuring the quality of service for computing tasks.

10 pages, 7 figures, 4 tables. Submitted to Proceedings of the CSEE. This is a bilingual manuscript. The English version appears first, followed by the Chinese version