Applying information extraction for abstracting and automating cli-based configuration of network devices in heterogeneous environments

Abstract

ith the continuous growth of current networks, configuration management has become increasingly relevant to the Information and Communication Technologies (ICT) field. Despite numerous standardization efforts, network administrators continue to rely on Command-Line Interfaces (CLIs) to modify and control the configuration of network devices. Nevertheless, network administrators must deal with the complexities that derive from this practice. On one hand, CLIbased configuration hinders the automation of network configuration tasks which are typically required in autonomic management. The only means for achieving a certain degree of automation is the creation of custom scripts, which is neither scalable nor practical, and is the reason why configuration management tasks are mainly performed through manual intervention. On the other hand, CLIs are generally both device and vendor-specific. In the context of heterogeneous network infrastructures—i.e., networks typically composed of multiple devices from different vendors—the use of several CLIs raises serious Operation, Administration and Management (OAM) issues. Moreover, multi-vendor configurations not only differ syntactically. Overall, the utilization of proprietary mechanisms allows neitherreusing the configurations nor sharing knowledge consistently between vendors’ domains. Due to this heterogeneity, CLIs typically provide a help feature which is in turn a useful source of knowledge to enable semantic interpretation of a configuration space. The large amount of information a network administrator must learn and manage makes Information Extraction (IE) and other forms of natural language analysis of the Artificial Intelligence (AI) field key enablers for the network device configuration space. In this chapter we present an Ontology-Based Information Extraction (OBIE) System from the Command-Line Interface (CLI) of network devices. This system exploits natural language resources already available in CLIs in order to extract relevant information and automatically build the semantics of each configuration space. Overall, our solution provides network administrators with a simple tool which entirely automates and abstracts the complexities and heterogeneity of underlying configuration environments in order to reduce time and effort in the configuration of network devices. With such a tool, network administrators will no longer have to read hundreds of manuals, and configuration scripts can be automatically updated for new devices or system upgrades. We developed a prototype implementation to show how we complete the loop from the process of IE, to the configuration of network devices and final testing.