Senior Network Administrator (Rosebank, Cape Town ) at Council for Scientific and Industrial Research (CSIR)

About the job:

The CSIR has a vacancy for a Senior Network Administrator in the Centre for High Performance Computing (CHPC) area, within the National Integration Cyberinfrastructure System (NICIS), which is hosted by Next Generation Enterprises and Institution cluster. The incumbent will be responsible to manage both the HPC and Cloud Network and other CHPC IT systems. This will include but not limited to, identify, analyse, and react to security incidents, events, and threats, and manage security controls. This position is based in Rosebank, Cape Town.

Key responsibilities:

• Design and implement network infrastructure for HPC and OpenStack Cloud, including physical and virtual network components;
• Manage HPC network and ensuring compliance with security policies and procedures;
• Implement network automation and orchestration using tools such as Ansible, Puppet, or Chef;
• Collaborate with OpenStack administrators, HPC administrators, developers, and operations teams to ensure smooth integration and operation of the OpenStack and HPC platforms;
• Keep up to date with the latest network technologies and trends, and evaluating their potential impact on HPC and OpenStack;
• Design and implement security measures, including firewalls, intrusion detection systems, and encryption technologies, to prevent unauthorized access and ensure the confidentiality, integrity, and availability of the HPC and OpenStack Cloud resources;
• Design and implement the network infrastructure to ensure optimal performance of the HPC systems. This includes configuring the HPC network switches, routers, and other hardware components to support the high-bandwidth requirements of the HPC applications;
• Monitor the performance of the HPC systems and identifies bottlenecks and other issues that affect system performance, optimize the performance of the system, including tuning the network and adjusting system settings;
• Provide technical support to users of the HPC systems, including troubleshooting and resolving issues related to system performance, and Network related issues;
• Ensure HPC network device cabling and connectivity needs are designed, implemented and deployed in keeping with industry standards;
• Work with and keep up to date on networking technology from a diverse range of vendors.

Qualifications, skills and experience:

• A Bachelor’s degree in information technology, information systems, computer science/engineering, electronic engineering, or related field with at least five years’ experience in an academic or research community environment, HPC network and Security support;
• Experience with system management, monitoring/alerting tools (e.g., Ganglia, Nagios, Prometheus, Zabbix)
• Experience in interfacing with HPC network technologies such as InfiniBand;
• Experience supporting (HPC) High Performance Computing environments;
• Experience with the use of configuration management and orchestration tools such as Ansible;
• Demonstrated experience with installation, configuration, testing and operation of network equipment from multiple equipment providers;
• Network programming experience including the ability to automate network management;
• Knowledge of Linux Open-source tools, operating systems, and network administration;
• Knowledge and experience diagnosing system issues on Linux-based clusters;
• Knowledge and experience with open Networking, specifically Cumulus Linux;
• Technical knowledge of Linux-based cloud infrastructure technologies, such as: virtualisation, containerisation, software-defined networking (SDN), network function virtualisation (NFV), high speed networks, and orchestration;
• Demonstrated ability to quantify, analyse, determine root cause, and resolve system and communication network issues, and develop preventive actions;
• Ability to work independently as well as collaboratively within a team, to include the ability to lead moderately complex projects or small project teams.

Closing date: 12 March 2024