The Bianca Cluster

Bianca is a research system dedicated for analyzing sensitive personal data. Bianca is part of the SNIC-SENS project.

Bianca provides each project with a virtualized "UPPMAX cluster environment" utilizing OpenStack as virtualization platform.

Bianca provides 3264 cores in the form of 200 dual CPU (Intel Xeon E5-2630 v3) Huawei XH620 V3 nodes with 128GB memory, eight "fat" nodes have 256 GB or 512 GB of memory, and four nodes with two NVIDIA A100 40GB GPUs each.

Reference datasets

We have the SweGen/1000 Genomes dataset as well as UKBB institutional genetics dataset available, as well as other common datasets. See accessing swegen and our data page for details for requesting access.


Security

Since Bianca is designed to handle sensitive personal data security is a key aspect of the configuration. In order to ensure that the data is safe we have implemented a series of security measures including, but not limited to:

  • One virtualized cluster per project, no resources are shared between projects.
  • Separate storage volumes per project.
  • Detailed logging of file transfers in and out of the cluster.
  • Two factor authentication
  • No internet access inside the clusters.
  • Locked racks for the hardware
  • Destruction of broken hard drives

Uppsala University has decided on the following KRT classifications for Bianca:

  • 321 for project directories
  • 322 for home directories

Technical Summary

  • 204 compute nodes with single or dual CPUs and one 4TB mechanical drive or 1TB SSD
  • Each CPU has 8 cores
  • 4 compute nodes each equipped with 2xNVIDIA A100 (40GB) GPUs
  • Total number of cores is 3264
  • Dual 10 Gigabit Ethernet for all nodes
Last modified: 2022-02-22