Our aim is to get the most out of supercomputing to provide solutions to yours engineering and scientific challenges.
Barcelona, January 10th 2016 -- Osakidetza-SVS selects the proposal submitted by Omega Peripherals and HPCNow! for the supply and installation of an HPC infrastructure that will equip the Unidad de Biomedicina Cuantitativa with the resources needed to provide computational services in a scientific environment.
The Unidad de Biomedicina Cuantitativa, linked to the Hospital Universitario Cruces (HUC)-BioCruces, provides to the investigators the tools needed to conduct activities such as signal processing, treatment of biomedical images, numerical simulation of biophysical and biodynamic processes, and the quantitative analysis of physiological temporal series. The proposal submitted by Omega Peripherals and the services provided by HPCNow! will allow the setup of a new cluster in order to support the clinical research carried out at HUC-BioCruces.
The solution is equipped with a total of eight ASUS servers with Intel Xeon Haswell processors, two of which will be used to manage the cluster. The administration nodes are equipped with 32 GB of memory and two 6 core Intel Xeon Haswell E5-2609 v3 processor. Four of the servers will act as computing nodes, and are equipped with 96 GB of memory and two 10 core Intel Xeon Haswell E5-2660 v3 processors. Finally, the two remaining servers will be used as storage BeeGFS nodes and will provide a total space of 24 TB of storage.
The software installed on the management nodes is a is a distribution of the sNow! Suite in its High Availability mode, developed by the HPCNow! team and specifically designed to manage clusters in scientific computing environments. The purpose of sNow! HA is to provide scientists and engineers with an easy to use and manage system, avoiding the complexities that come with the management of computer clusters.
Regarding the storage nodes, they are unified with BeeGFS, a parallel filesystem focused on three key concepts: scalability, good usability and flexibility. With BeeGFS, increasing the number of servers or disks in the system can scale bandwidth, storage or metadata.
In order to put the new infrastructure into operation, the HPCNow! team was in charge of both the installation of the hardware that integrates the solution and server configuration, including the installation of the sNow! suite HA and the BeeGFS parallel filesystem, as well as testing and optimizing the applications that will run on the cluster.