What is MARVIN cluster?
MARVIN cluster is a system where users can send their processes to be computed and each lab has queue size according to their necessity
MARVIN cluster is a set of interconnected machines (nodes). Each node has several cores (CPUs).
Every node is sharing information with other nodes whithin a common shared space.
It is possible to parallelize jobs within the same node (along all the CPUs) or between nodes (using CPUs from both nodes).
- marvin.s.upf.edu frontend and login node.
- NFS server node with 180TB.
- All nodes connect to IBM Spectrum Scale with Infiniband (56Gbs).
- 880 TB Parallel FIle System (IBM Spectrum Scale - GPFS)
- The OS running in the computing nodes is CentOS 7.8 x86_64
- 12 computing nodes with 16 cores and 128 GB RAM.
- 13 computing node with 32 cores and 256 GB RAM.
- 3 computing node with 48 cores and 512 GB RAM.
- 1 computing node with 64 cores and 512 GB RAM.
- 2 computing node with 48 cores, 1.5 TB RAM and GPU.
Our system has a total of 30 computing nodes, which all in all sum 912 cores. And the disk space available for MARVIN and the nodes are approximately 9.984 TB RAM. The Global Disk Status is divided into three main groups:
homes: 20 TB
scratch: 500 TB
projects: 300 TB
All of them GPFS. Thus meaning high performance access.
Download the Marvin use Cheat Sheet.