Salary: £48,607 – £58,528 plus benefits
The Research Computing Service operates a large, heterogenous cluster tailored to support the diverse computing needs of the research community, with high-throughput capacity, high-end capability, GPU compute and large-memory jobs. Currently the service has enrolled over 4600 users and there are 1.1 million jobs run each week.
This role involves significant amounts of systems development work, including systems analysis and implementation. The evaluation and integration of new technology plays a major part in HPC provisioning.
Duties and responsibilities
- Design, implement, and support large commodity clusters and large shared memory systems.
- Support and assist research groups in running their code.
- Provide operational support to investigate, define, and resolve complex HPC issues.
- Identify, discuss, and implement continuous improvements.
- Maintain knowledge of industry trends and technologies, working closely with the specialists of leading vendors.
- Help and train more junior colleagues, sharing experience and knowledge.
- Fully understand the importance of security and seek specialist security advice to support implementation.
- Engage with a complex community of stakeholders, including senior technologists, thought leaders and business leaders from ICT and the wider College, facilitating collaboration.
- Degree or equivalent, preferably with a technical qualification or comparable
- In depth working experience of HPC system architecture, both proprietary and commodity clustered computing solutions, including current interconnect technologies such as gigabit Ethernet and InfiniBand
- Knowledge of managing large-scale high throughput storage solutions
- Practical experience of using Linux and HPC clustered computing software stacks such as imaging, monitoring, parallel command execution and scheduling/queuing solutions
- Knowledge and experience of debugging applications using strace and core dumps
- Knowledge and experience of using one or more scripting languages such as sh, python.
- Experience of MPI library implementation and knowledge of parallel methodologies.
- Problem analysis followed by proposal and implementation of resolutions
- Knowledge of parallel programming methodologies
- Experience of producing documentation, web pages and training materials
- Experience with the PBS Pro scheduler and CentOS 8.
- Excellent interpersonal skills with the ability to lead others and persuasively communicate complex technical information effectively to a diverse range of stakeholders, including those without a technical background
- Ability to identify good testing/evaluation methods for problem solving on operational platforms
- Ability to evaluate hardware and software technologies and to select those which best fit the requirements of the HPC service
- Ability to plan and undertake individual work activities within general guidelines, using initiative and making decisions. Understanding when to bring in knowledge or expertise from other groups.
- Ability to write reports
- Awareness of cost and budgets, working within budgetary constraints, suggesting alternatives as appropriate.
- Experience of working in a research/technology environment and working in a customer-oriented environment
- Where packaged solutions are not available, ability to propose and develop bespoke components to provide the features required by the research community
The role is an open ended contract, at Level 4.
Please note that job descriptions cannot be exhaustive and the post-holder may be required to undertake other duties, which are broadly in line with the above key responsibilities.
Imperial College is committed to equality of opportunity and to eliminating discrimination. All employees are expected to follow the 7 Imperial Expectations detailed below:
- Champion a positive approach to change and opportunity
- Encourage inclusive participation and eliminate discrimination
- Communicate regularly and effectively within and across teams
- Consider the thoughts and expectations of others
- Deliver positive outcomes
- Develop and grow skills and expertise
- Work in a planned and managed way
Should you require any further details on the role please contact: email@example.com.