The rise of containerization has led to the development of container cloud technology, which offers container deployment and management services. However, scheduling a large number of containers efficiently remains a significant challenge for container cloud service platforms. Traditional load prediction methods and scheduling algorithms do not fully consider interdependencies between containers or fine-grained resource scheduling, leading to poor resource utilization and scheduling efficiency. To address these challenges, this paper proposes a new load prediction model CNN-BiGRU-Attention and a container scheduling strategy based on load prediction. The prediction model CNN and BiGRU focus on the local features of load data and long sequence dependencies, respectively, as well as introduce the attention mechanism to make the model more easily capture the features of long distance dependencies in the sequence. A container scheduling strategy based on load prediction is also designed, which first uses the load prediction model to predict the load state and then generates a scheduling strategy based on the load prediction value to determine the change of the number of container replicas in a fine-grained manner based on the load prediction value in the next time window, while the established domain-based container selection method is employed to facilitate the coarse-grained online migration of containers. Experiments conducted using public datasets and open-source simulation platforms demonstrate that the proposed approach achieves a 37.4% improvement in container load prediction accuracy and a 21.7% improvement in container scheduling efficiency compared to traditional methods. These results highlight the effectiveness of the proposed approach in addressing the challenges faced by container cloud service platforms.

Due to its development agility, continuous delivery, scalability and other characteristics, the microservice architecture systems (MASs) have provided complex business functions to hundreds of millions of users in many application fields. The operation and maintenance governance for a large number of microservices with complex relationships is crucial to ensuring the stability and reliability of an MAS. Although this research field has received certain attention and produced some innovative results, there is a lack of systematic reviews covering the different aspects of it. In this context, the central objective of this study is to carry out a systematic literature review (SLR) in this field, in an attempt to review existing issues, discuss the main trends, and share the findings with the academia. As a result, we start from more than 500 scientific papers published from 2009 to 2021 and extract 144 most significant papers, identify that the main research directions of this field include load balancing, fault detection, and autoscaling. Subsequently, we provide a comprehensive description of these research directions, discuss them in particular detail. We also determine limitations of current work and discuss new directions worth exploring in the future. Consequently, the outcomes will assist professionals and experts in the industry as well as academic researchers to focus more on operation and maintenance governance of MASs and further improve the relevant methods and theoretical systems in this field.

The resource release bugs are a common type of serious programming bug. However, it is hard to catch them by using static detection for the lacking of comprehensive prior knowledge about the release functions. In this paper, a resource release bug detection method is proposed by introducing analogical reasoning on word vectors. First, the functions of the target source code are encoded into word vectors by the word embedding technique in natural language processing. Second, a two-stage reasoning method is developed for automatically identifying unknown resource release functions according to a few well-known seed functions. 3CosAvg algorithm is employed for the first stage, and a new algorithm is designed for the latter, called 3CosAddExchange. Finally, the identified release functions are translated into static analysis rules to detect potential bugs. The experiment shows that the proposed method is effective and efficient for the large-scale software project. Five unknown resource release bugs are successfully detected in the Linux kernel and confirmed by kernel developers.