3.3 Prerequisites for First Use
Before starting, prepare BioLine access, fkit, input data, and a host environment that meets the selected Tool’s resource requirements.
- An upstream Flow requires a valid BioLine account and a readable data location. fkit manages credentials; do not write them into task directories.
- Downstream tasks require a compatible local container runtime that can run offline, mount stage read-only, and write task outputs.
- Local storage is needed for the exact Tool version pulled for this task, staged inputs, and results. The historical image set does not need to be downloaded in advance.
- CPU, memory, disk, and GPU requirements come from the selected Tool’s config/summary and the host runtime. Use a GPU only when that exact Tool supports it and the host is compatible.
- Input formats depend on the objective and the discovered Tool/Flow interface. Common inputs include paired-end FASTQ, abundance or pathway tables, ASV tables, and metadata.tsv.
| Requirement | Purpose |
|---|---|
| fkit and a BioLine account | Discover, inspect, and run platform Flows, Jobs, and Tools. |
| Container runtime | Run downstream task scripts within the restricted local boundary. |
| Available disk and compute | Hold this task’s exact Tool version, inputs, and outputs. |
There is no need to keep a fixed historical image set locally. Discover existing BioLine Tools and Flows first, then inspect the exact version, documentation, and resource configuration. If no suitable capability exists, create one on demand.