Albedo will be the successor of Ollie. NEC has won the procurement and will install in June 2022 (rather late due to the general delays in delivery ):
- 2 login nodes
- 240 compute nodes with
- 2x AMD Rome Epyc 7702 2GHz (3.3GHz Boost), 64 Core, cTDP reduced from 200W to 165W
- 256 GB RAM,
- 500GB SSD
- Fast interconnect: HDR Infinband
- 4 "fat" nodes as above, 4TB RAM, 500GB SSD + 7.5TB SSD
- 1 GPU node with 1TB RAM, 4x NVIDIA A100/80
- + one additional GPU node with smaller equipment for interactive access: testing, Jupyter notebooks,...
- More GPU nodes will follow later, after we gained first experience of what you really need, and to offer most recent hardware
- Our small test node with NEC's new vector engine "SX-Aurora TSUBASA" can be integrated
- 5 PB /scratch NEC GxFS (IBM Spectrum Scale)
- 220 TB as NVMe SSDs as fast cache and/or burst buffer
- extension (capacity, bandwidth) possible
- All nodes connected to /isibhv (NFS, 10GbE)
- Alma Linux ("free RedHat")
is the tier-3 High Performance Computing platform (HPC) hosted and supported at AWI. In this documentation you can find the basics on how to operat it. Please, be aware that a basic knowledge on linux and HPCs is expected from Albedo's users, so this documentation does not cover all there is to know about HPCs, linux user permissions, data management...
That being said, Albedo documentation is a living document and we will do our best to improve it and answer the most common questions of the users, either with expanding the documentation or pointing at external sources.
| Children Display |
|---|