Cluster Storage Map

Not every data store is available everywhere on the cluster, nor is the filesystem performance equal for all nodes. The differences depend on which island a node is a member of and whether it is a login or a compute node. General CPU node islands are called “Emmy Phase X” where X indicates the hardware generation (1, 2, 3, …). General GPU node islands are called “Grete Phase X” where X indicates the hardware generation (1, 2, 3, …). Other islands exist for specific institutions/groups or historical reasons (e.g. SCC Legacy). Which systems can be accessed and their relative performance of the link between each group/island of nodes (logins and partitions) and each data store is shown in the diagram below

Diagram of the connections between each HPC node group and the storage systems. The SCC Legacy login nodes (gwdu[101-102,108]) have a very slow connection to the GWDG ARCHIVE (Stornext AHOME). The MDC login nodes (Emmy Phase 2, glogin[3-8]) and RZG login nodes (Grete and Emmy Phase 3, glogin[9-13]) have a very slow connection to PERM. All nodes have a slow-medium connection to the HOME/Project storage (VAST), Software, and the Project Map. The SCC Legacy login nodes (gwdu[101-102,108]) and SCC Legacy compute nodes have a slower-medium connection to the GWDG Unix HOME (Stornext) and a very fast connection to the SCRATCH SCC (BeeGFS). The MDC login nodes (Emmy Phase 2, glogin[3-8]) and Emmy Phase 2 compute nodes have a very fast connection to the SCRATCH MDC (formerly known as SCRATCH Emmy), a medium connection to the SCRATCH RZG (formerly known as SCRATCH Grete), and a slower-medium connection to the SCRATCH SCC (BeeGFS). The RZG login nodes (Grete and Emmy Phase 3,glogin[9-13]), Grete compute nodes, and Emmy Phase 3 compute nodes have a very fast connection to the SCRATCH RZG (Lustre), a medium connection to the SCRATCH MDC (Lustre), and a slower-medium connection to the SCRATCH SCC (BeeGFS).

HPC Storage Systems

Connections between each HPC node group/island and the different storage systems, with the arrow style indicating the performance (see key at bottom right).

There are a few important points to note:

  • Portal SCC users have access to some CPU and GPU nodes in the Grete Phase 2 & 3 and Emmy Phase 3 islands (the scc-cpu and scc-gpu partitions) in addition to the SCC Legacy island.
  • Legacy SCC users only have access to the SCC Legacy island and have their HOME directories on the GWDG Unix HOME filesystem.
  • Each island (SCC Legacy, Emmy Phase 2, Grete & Emmy Phase 3) has a separate SCRATCH/WORK data store connected via a very fast network, with slower connections to the other ones.
  • ARCHIVE/PERM data stores are only accessible from login nodes.
  • Temporary Storage for Slurm jobs is always allocated in RAM, on fast local SSDs, and the fastest SCRATCH/WORK available for the node.
  • The CIDBN, FG, and SOE islands are not shown in the diagram above but have access to the same storage systems with the same relative performace as the SCC Legacy, though some also have their own dedicated SCRATCH/WORK.
Info

See CPU Partitions and GPU Partitions for the available partitions in each island for each kind of account.

See Logging In for the best login nodes for each island (other login nodes will often work, but may have access to different storage systems and their hardware will be less of a match).

See Software Stacks for the available and default software stacks for each island.

See Types of User Accounts if you are unsure what kind of account you have.