High Performance Computing
Key research areas
Read pre-preprocessing, De novo assembly, Gene annotation, Variant discovery, Mapping, Differential Expression, Orthology analysis, Phylogenomics, Population genomics, Metabarcoding, Βiodiversity index calculation, Εcological data analysis
“Zorbas” is IMBBC’s High Performance Computing (HPC) system, dedicated to bioinformatics applications for non-model species and ecological data analyses. Equipped with ~400 cores, >2TB RAM, 640GB of which is available on a single node, “Zorbas” hosts more than 200 state-of-the-art software suites. Among other supported types of analysis include: non-model organism next-generation sequencing data, environmental -omics, and up-to global scale biodiversity and ecology data crunching. Web-based interfaces like the RvLab (https://rvlab.portal.lifewatchgreece.eu) provide user-friendly and seamless access to the HPC system. Upon request command-line access is also possible (https://hpc.hcmr.gr/docs/getting-started/).
For the technology-literate: ongoing efforts translate a more-than-a-decade long of bioinformatics data processing in containerized pieces of software (Singularity and Docker based) as well as reusable pipelines (in snakemake).
“Where can I find more info?” The dedicated IMBBC HPC portal https://hpc.hcmr.gr/ contains the most up-to-date information about the IMBBC HPC system on how to make the best use of it for your research purposes. Detailed documentation covering basic to advanced topics of working in an HPC is available (https://hpc.hcmr.gr/docs/). Logistics on how to gain access to the IMBBC HPC is described at: https://hpc.hcmr.gr/docs/getting-started/
“Which type of analyses have been supported by IMBBC HPC?” Analyses at IMBBC HPC cover most of the -omics levels from DNA (such as genomics and metabarcoding) and RNA (transcriptomics) to phenomics and community ecology. Practical descriptions and publications achieved through the use of the IMBBC HPC are available at: https://hpc.hcmr.gr/use-cases/ and https://hpc.hcmr.gr/publications/
In addition a dedicated helpdesk service is available https://helpdesk-hpc.hcmr.gr/ to address software and hardware requests as well as maintenance operations. A Resources Booking calendar https://booking-hpc.hcmr.gr/ can help users to pre-register their analyses. A real-time monitoring site: http://ganglia.her.hcmr.gr/ is there to assist users in synchronising their jobs with the jobs of others.
Is there a one page summary of the IMBBC HPC system? An in-a-nutshell poster describing the IMBBC HPC system is available at: https://hpc.hcmr.gr/?attachment_id=2251 (presented at the Hellenic Bioinformatics conference 2019)
IMBBC HPC (“Zorbas”) in numbers:
- 19 worker nodes/4 computing partitions
- 328 Intel Xeon cores
- 2.3TB total RAM
- 640GB RAM on a single node
- 1.5TB RAM on a single node in 2020
- 40Gbps Infiniband interconnection
- 11 Tflops peak performance
- Read pre-preprocessing
- De novo assembly
- Gene annotation
- Variant discovery
- Differential Expression
- Orthology analysis
- Population genomics
- Βiodiversity index calculation
- Εcological data analysis
- > 50 users (Greece, Italy, Spain etc)
- 2 HPC system administrators specialized in bioinformatics applications
- ~20000 submitted jobs in 2019
Ongoing Collaborations: Ongoing collaborations with the LifeWatch ERIC, EMBRC ERIC, and the ELIXIR Research Infrastructures aim at sharing resources and accumulated expertise to the broader pertinent Hellenic and European bioinformatics and biology communities.
Funding: The establishment of Zorbas was funded by the MARBIGEN (EU REGPOT) project, Lifewatch Greece, CMBR (Centre for the study and sustainable exploitation of Marine Biological Resources) and Elixir Greece Research Infrastructures (ESFRIs)
- Potirakis A., et al. ZORBAS: the HPC cluster for biological analysis in IMBBC, HCMR. Poster presented at: 12th Hellenic Bioinformatics Conference; 2019 Oct 11-13; Heraklion, Crete, Greece
- Documentation for IMBBC’s High Performance Infrastructure: https://hpc.hcmr.gr/docs/