Problem
Research questions and friction points this paper is trying to address.
Electromigration
Manycore Systems
Degradation-Balanced Scheduling
Task Placement
Reliability
On the Impact of Degradation-Balanced Scheduling in Manycore Systems
🤖 AI Summary
This study addresses the imbalanced electromigration degradation in multi-core systems caused by scheduling strategies that rely solely on instantaneous reliability states. To mitigate this, a Reliability Balancing (RB) strategy is proposed. Its core innovation lies in shifting from static state evaluation to dynamic, task-wide effect prediction by modeling the combined impact of power, temperature, and current density on degradation. This predictive framework is integrated with a multidimensional penalty mechanism and multi-objective optimization to enable intelligent scheduling. Experimental evaluations under randomized mixed workloads demonstrate that the proposed approach improves average and minimum system reliability by 0.37% and 4.35%, respectively, while reducing the reliability standard deviation by 10.84%. These results indicate that the RB strategy significantly extends system lifetime and enhances operational stability.
📝 Abstract
Electromigration degradation in manycore systems depends strongly on how runtime workload is distributed across cores. Task placement affects power, temperature, and current density, which in turn shape the spatial distribution of degradation over time. However, a scheduler that ranks cores only by their current reliability state may not always distinguish among candidate cores, especially when reliability values are close at decision time. In this case, scheduling decisions can be driven by secondary factors rather than by long-term degradation balance. This paper studies degradation-balanced scheduling and proposes a Reliability-Balanced (RB) policy that predicts the full-task effect of each candidate assignment. RB penalizes added imbalance in electromigration exposure, stress, active time, and task count while guarding the predicted Rvalue floor. On randomized continuous mixed workloads, the proposed scheduler preserves completed tasks while improving average Rvalue by 0.37% and minimum Rvalue by 4.35%. It reduces Rvalue, exposure, stress, and active-time standard deviation by 10.84%, 9.39%, 8.35%, and 16.38%, respectively.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Degradation-Balanced Scheduling
Electromigration
Manycore Systems
Reliability-Balanced Policy
Task Placement
🔎 Similar Papers
No similar papers found.
