Future situation:
Unlike the classic HPC centres, which are primarily intended for intensive computing, HPC RIVR will also enable the processing of a large amount of data and rapid exchange with other computing centres as well as computing on modern GPUs. The central HPC RIVR supercomputing system with a computing capacity of at least 1 petaFLOP/s will be placed in the premises of IZUM. For development activities and testing of system and programme solutions, the University of Maribor will establish a prototype HPC RIVR system with a computing capacity of at least 220 teraFLOP/s. The HPC RIVR @ IZUM system will provide a permanently available capacity of computing, processing and storing data for all users, while the HPC RIVR @ UM system will enable numerous possibilities for reconfiguration and testing of various HPC solutions without threatening the stability of the core node HPC RIVR @ IZUM. Evidently better and stable solutions will be periodically transferred between the systems, which will be connected with a powerful high throughput network equipment (at least 100 Gbps).