共 24 条
In-depth analysis of Kubernetes manifest verification tools for robust CNF deployment
被引:0
|作者:
Diarra, Boubacar
[1
,2
]
Guillouard, Karine
[1
]
Ouzzif, Meryem
[1
]
Merle, Philippe
[2
]
Stefani, Jean-Bernard
[3
]
机构:
[1] Orange Innovat Rennes, Cesson Sevigne, France
[2] Inria Lille, Villeneuve Dascq, France
[3] Inria Grenoble Alpes, Montbonnot St Martin, France
来源:
PROCEEDINGS OF THE 27TH CONFERENCE ON INNOVATION IN CLOUDS, INTERNET AND NETWORKS, ICIN
|
2024年
关键词:
Kubernetes;
manifests verification;
verification tools;
evaluation criteria;
CNF;
robust CNF deployment;
D O I:
10.1109/ICIN60470.2024.10494445
中图分类号:
TP18 [人工智能理论];
学科分类号:
081104 ;
0812 ;
0835 ;
1405 ;
摘要:
Kubernetes has emerged as the leading platform for orchestrating cloud-native network functions (CNF). The description of the hardware and software architecture of systems to be deployed with Kubernetes is defined in configuration files called manifests. In order to address the errors that these manifests may contain, manifest verification tools have been developed. However, evaluating and selecting the most appropriate verification tool for a given project is a difficult task. To facilitate this evaluation process, we identified and characterized the most relevant verification capabilities, useful also in the CNF context. The selected capabilities serve as criteria for the evaluation of manifest verification tools. In total, we have identified 13 criteria, like the variety of verification types or the ability to carry out JSON Schema-based verification. The identified criteria were used to evaluate 16 of the most commonly used verification tools in the Kubernetes community. This evaluation revealed several shortcomings in the existing tools.
引用
收藏
页码:17 / 24
页数:8
相关论文