Energy harvesting Internet of Things (IoT) devices are capable of sensing only intermittent and coarse-grained data due to sleep scheduling; therefore, we develop a restoration mechanism (e.g., probabilistic matrix factorization (PMF)) that exploits spatial and temporal correlations of data to build up an environmental monitoring system. However, even with a well-designed restoration mechanism, a high accuracy of the environmental map cannot be achieved if an appropriate sleep scheduling of IoT devices is not incorporated (e.g., if IoT devices at necessary locations are in sleep mode or are not involved in restoration due to their insufficient energy). In this paper, we propose a restoration-aware sleep scheduling (RASS) framework for energy harvesting IoT-based environmental monitoring systems. Here, RASS involves customized deep reinforcement learning (DRL) considering the restoration mechanism, using which the controller performs sleep scheduling to achieve high accuracy of the restored environmental map while avoiding energy outage of IoT devices. The evaluation results demonstrate that RASS can achieve an environmental map with 5% or a lower difference from the actual values and fair energy consumption among IoT devices.

Training on time-series data generated from mobile networks is a resource-intensive and time-consuming task that encounters various training failures. To cope with this issue, we propose CheckBullet, a lightweight checkpoint system to minimize storage requirements and enable fast recovery in mobile networks. First, CheckBullet determines a checkpointing interval based on the characteristics of the model and the timing of failure occurrences. This approach ensures fast recovery while preserving the existing training runtime. Second, CheckBullet quantizes the weight tensor and eliminates duplicate weights, which significantly reduces the overall checkpoint size, leading to a substantial decrease in storage requirements. Third, CheckBullet selects the minimum training loss among the deduplicated checkpoints and merges the selected checkpoints. This approach reduces recovery time while preserving existing training loss. The experimental results show that CheckBullet can reduce the recovery time by$6\times$to$11\times$barely increasing the training runtime. Furthermore, CheckBullet can save storage requirements by up to 70% while maintaining the minimum training loss.