A hardware and software computational platform for HiPerDNO (High Performance Distribution Network Operation) project

The HiPerDNO project aims to develop new applications to enhance the operational capabilities of Distribution Network Operators (DNO). Their delivery requires an advanced computational strategy. This paper describes a High Performance Computing (HPC) platform developed for these applications whilst also being flexible enough to accommodate new ones emerging from the gradual introduction of smart metering in the Low Voltage (LV) networks (AMI: Advanced Metering Infrastructures). Security and reliability requirements for both data and computations are very stringent. Our proposed architecture would allow the deployment of computations and data access as services, thus achieving independence on the actual hardware and software technologies deployed, and hardening against malicious as well as accidental corruptions. Cost containment and reliance on proven technologies are also of paramount importance to DNOs. We suggest an architecture that fulfills these needs, which includes the following components for the HPC and Data Storage systems: • Hadoop Distributed File System • a federation of loosely coupled computational clusters • the PELICAN computational application framework