When a research project begins, the first need is to know where the data will be stored.
But it’s important to look beyond this first question to understand the real need(s):
- What do I need to store?
- Scientific data, codes, documents?
- Binary or ascii files?
- What is/will be the volume of data?
- What level of security/backup is required (can the data be easily reproduced?)
- What will I do with the data?
- Processing or calculation
- Analysis, mining
- Sharing
- Preservation …
- Is the data sensitive or confidential?
- Who will need to access the data?
- In what way will the data be accessed?
- What are the data flows going to be (continuous, regular, on demand, etc.)?
- When will I need to access the data?
- Quickly, regularly
- In a few months
- Maybe in a few years
- At the end of the project, what volume of data will need to be preserved and for how long?
- What funding is envisaged for data management and storage?
This final question is not a technical one, but it does have consequences that can be significant depending on the technical choices made. Data storage (in the wider sense) has a material and human cost that must be taken into account when research projects are submitted.
There are a few definitions and concepts that are important to understand before making choices and choosing the services available.
