Skip to content

Virtual Infrastructure University of Cordoba

Contact Infrastructure Provider

Summary

Description

The University of Córdoba has a virtual infrastructure useful for the testing of products or services in the agri-food sector, characterized by multi-layered connectivity ranging from low power to 5G, in addition to an advanced data platform.

Besides, the UCO infrastructure includes the following components:

Supercomputing:

A massive memory supercomputer capable of managing 12TB DRAM in a single process, thanks to 8 interconnected Intel Xeon Cascade Lake 8260M processors, which includes:

  • 1 bull X808 server.

  • 192 cores distributed across 8 Intel Xeon Cascade Lake 8260M processors, each with 24 cores at 2.4 GHz.

  • 12 TB of memory in 96 DIMMs of 128 GB DDR4 at 2666 MHZ.

  • 2 SSD disks of 400 GB for the operating system.

  • 1 infiniband EDR card of 100 Gbit.

  • 8 ports of 10 Gigabit SFP+ and 16 Gbit, plus 1 IPMI for administration.

A parallel computing cluster with 960 cores, detailed as follows:

  • 6 bull X440 chassis hosting 960 cores.

  • 24 computing nodes, each with:

  • 2 Intel Xeon Cascade Lake 6230 processors, each with 20 cores at 2.1 GHz.

  • 192 GB of RAM distributed in 12 DIMMs of 16 GB DDR4 at 2933 Mhz.

  • 1 SSD disk of 240 GB for the operating system, with up to 3 disks possible per node.

  • 2 disks of 4 TB SATA3.

  • 1 infiniband EDR card and 2 ports of 10 Gigabit.

NVIDIA DGX2 server, specifications:

  • GPUs: 16 NVIDIA® Tesla V100.

  • GPU Memory: 512GB total.

  • Performance: 2 petaFLOPS.

  • NVIDIA CUDA® Cores: 81920.

  • NVIDIA Tensor Cores: 10240.

  • NVSwitches: 12.

  • Maximum Power Usage: 10kW.

  • CPU: Dual Intel Xeon Platinum 8168, 2.7 GHz, 24 cores.

  • System Memory: 1.5TB.

  • Infiniband EDR 100Gb.

  • Dual 10 GbE.

  • Storage OS: 2X 960GB NVME SSDs.

  • Internal Storage: 30TB (8X 3.84TB) NVME SSDs.

Laboratory of the Research Group GIIA, equipped with the latest generation GPGPUs:

  • NVIDIA JETSON AGX ORIN.

  • 4 JETSON NX ORIN.

  • 2 JETSON NANO ORIN.

  • Other RTX, Nano, NX, TK GPU models

  • Xilinx FPGAs