Sections

you are here you are here: home següent Service file


2 images available
JFF Parallel computing cluster

Service consisting in computational parallel cluster formed by 128 nodes and 1024 cores, deployed in 4 selves having each one 32 nodes, 8 cores each and 16 Gb RAM memory. The nodes are interconnected by a Infiniband high speed network, able to solve latencies of 2.25 microseconds, 20 Gb/s capacity. Parallel file system, called Lustre, allows unified capacities of several Petabytes (Pb), being highly scalable, and obtaining the optimal efficiency using a high performance hardware, with a useful maximum capacity of 1 Pb with 1200 disks of 1 terabyte (Tb), writing at 5.6 Gbytes/s. This unit has been acquired with 80 disks of 1 Tb, allowing a storage capacity of 64 Tb and writing at 2.6 Gb/s.     


Information and use

Scientific and Technical Services at the UPC
© UPC Open link in new window. Universitat Politècnica de Catalunya · BarcelonaTech.