2020, Vol.25, No.9

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

pISSN : 1598-849X / eISSN : 2383-9945
https://journal.kci.go.kr/jksci

In this paper, we propose a method of in-depth analysis of internal operation process by recording library calls and related information that occur in the operation process of NoSQL database. It observes and records the specified library calls, compares the internal behavior differences between the NoSQL databases through recorded library call information, and evaluates the characteristics and scalability of each database by observing changes in the number of input data. The development of computing performance and the activation of big data have led to the emergence of different types of NoSQL databases for recording and analyzing various and large amounts of data, and it is necessary to evaluate the scalability of each database in order to select a database suitable for each environment. However, it is difficult to analyze or predict how a database operates in traditional ways, such as benchmarking, observing external behavior through performance models, or analyzing structural features based on design. Therefore, it is necessary to utilize the techniques proposed in this paper to understand the scalability of NoSQL databases with high accuracy.

In this paper, we propose the possibility of using containers in the HPC ecosystem and the criteria for selecting a proper PMI library. Although demand for container has been growing rapidly in the HPC ecosystem, Docker container which is the most widely used has a potential security problem and is not suitable for the HPC. Therefore, several HPC containers have appeared to solve this problem and the chance of performance differences also emerged. For this reason, we measured the performance difference between each HPC container and Docker container through NAS Parallel Benchmark experiment and checked the effect of the type of PMI library. As a result, the HPC container and the Docker container showed almost the same performance as native, or in some cases, rather better performance was observed. In the result of comparison between PMI libraries showed that PMIx was not superior to PMI-2 in all conditions.

In this paper, we propose quantitative evaluation method that enable security comparison between Security Container Runtimes. security container runtime technologies have been developed to address security issues such as Container escape caused by containers sharing the host kernel. However, most literature provides only a analysis of the security of container technologies using rough metrics such as the number of available system calls, making it difficult to compare the secureness of container runtimes quantitatively. While the proposed model uses a new method of combining the degree of exposure of host system calls with various external vulnerability metrics. With the proposed technique, we measure and compare the security of runC (Docker default Runtime) and two representative Security Container Runtimes, gVisor, and Kata container.