The HERMES supercomputing cluster is the scientific infrastructure most used by I3A researchers. It responds to the increasing complexity of the simulations carried out in the most advanced research projects of the Institute.
HERMES has over 3260 cores and 5920 parallel processing threads, 26TB RAM, 264TB nvme and 400TB HDD shared storage, all connected by a 100Gbps backbone network. In addition we currently have 251,392 cuda cores, 720GB graphics memory, 1,179.42 TFLOPS peak FP16, and 828.6 TFLOPS peak FP32. HERMES is capable of processing a year's worth of work on a personal computer in as little as 3 hours, greatly accelerating the work of researchers. The I3A technical staff maintains the system and provides technical support, advice and training in parallel supercomputing and high-performance base coding.
The cluster is unit 27 of NANBIOSIS, the integrated infrastructure for the production and characterization of nanomaterials, biomaterials and systems in biomedicine, of the CIBER-BBN and the Jesús Usón Minimally Invasive Surgery Center, which has been recognized by MINECO as a Singular Scientific-Technological Infrastructure (ICTS).
Highlights:
Cores
Threads
RAM
NVMe
HDD
Gbps backbone network
CUDA cores
Graphic memory
