Welcome to Garnatxa’s documentation!
Garnatxa is the High Performance Computing (HPC) cluster located at Data Center of the Institute for Integrative Systems Biology (I2SysBio). The I2SysBio is a joint collaborative research institute involving University of Valencia (UV) and Spanish National Research Council (CSIC), open to the strategic involvement of biotech companies. The Research Center is placed in the Scientific Park (Parc Científic, Campus Burjassot-Paterna, Valencia, Spain).
In this users guide, we will go through the basics of using the HPC systems at I2SysBio. This includes logging into the systems , navigating the filesystem and learning how job submission works using the SLURM scheduler. If you need more help after this guide, please refer to the FAQ and User Guide pages. Additionally, you can also request help through the contact section.
Introduction
Garnatxa cluster is a collection of supercomputers (nodes), that communicate using a dedicated high performance network, making it an ideal location to scale computational analysis (especially in the biology environment) that otherwise would be impossible do. The High performance computing (HPC) cluster is free to use for the students, faculty and researchers working in I2SysBio. Garnatxa is equipped with about 464 cores (928 threads), 3 PB of distributed storage and 14TB of RAM, the cluster runs the SLURM scheduling manager. You can see more detailed information about Garnatxa in the hardware configuration section
Who manages Garnatxa?
We have high performance computing specialists managing the Garnatxa HPC cluster. HPC staff provides a variety of ways to help the researchers and students in computational tasks. The I2SysBio’s Data Center provides the cyber-infrastructure needed for cluster computing, while the compute resources are owned and expanded year after year through regional, national and European financed projects . It removes the stress of administering the large computational equipment from researchers and encourages them investing in communal resources.
Who Garnatxa is for?
If you have any computationally intensive tasks which need to be executed quickly or need an intensive I/O bandwidth and high storage requirements, then Garnatxa is the perfect place to run such tasks. If you belong to an external institution then contact us and we will inform you about the payment plans (public rates are published every year). You can use Garnatxa to perform the tasks like Theory, Bioinformatics and Computation, Genomics of Gene Expression, Theoretical and In Silico Modeling of Biological Systems, Computational Synthetic Biology, De Novo Synthetic Biology, Virus Experimental Evolution, Molecular Epidemiology, Bacterial Pathogenomics, Viral Biology, Environmental and Biomedical Virology, Evolutionary Systems Virology, Biosystems Design, Non-Coding RNA-Mediated Regulatory Networks, Transcriptional Orchestration of Metabolism, Panomics and Evolutionary Systems Microbiology, Evolutionary Genetics, Biotechnology and Synthetic Biology, Industrial Yeasts Biotechnology, Systems Metabolic Engineering and much more.
And now …
If you are interested in use Garnatxa you could start by reading the hardware configuration section and then quickstart guide. On those sections you will find more detailed instructions on how to request an account, how to connect to Garnatxa and finally how to submit your jobs to the system.
Also you can download a brief guide about procedures of use in Garnatxa
.
Quickstart
Storage system
Software and libraries
Working in Garnatxa
- Usage rules
- Interactive jobs
- Submitting jobs to Garnatxa
- System limits
- Priority policy
- Jobs in Garnatxa
- Basic concepts about SLURM
- Memory and CPU requests
- Slurm commands
- Submitting jobs to the cluster:
sbatch
- Basic, Single-Threaded Job
- Multi-Threaded SMP Job
- Message Passing Interface (MPI) Jobs
- Parallelization of data: ArrayJobs
- Parallelization of data using a file of commands (arrays version)
- Parallelization of data: srun in background (not recommended)
- Parallelization of data using a file of commands (background version - not recommended)
- Pipelines: Nextflow
Annexes