Host-an-Informatician: You Lab

Meetings

Meeting notes are in Google doc (private)

Summary of initial meeting

• Investigators are analyzing microscopy data. 30-50GB per experiment, some investigators analyzing 3 experiments per week
• Analysis is running on laptop computers - some personal assets
• Data storage is also on laptop computers or personal assets
• Data backup/archival does not seem robust, stored on consumer-grade hard drives
• Cost of storage resources is a concern
• Analysis code is primarily written in matlab
• Some past attempts to use cluster yielded differing results
• Unknown if version control is used
• Questions about how to access cluster or other resources
• Researchers present are not acquring/analzying human subject data

Informatics support opportunities

• Data storage on GCB resources
• Analysis on GCB resources
• Version control for analysis processes
• Provide training on existing infrastructure

Need to identify

• Storage/retention/archival requirements / possibilities. Do we support long-term storage?
• compute memory/processor resources.
• hosts to use? hardac? dscr? new vms? containers?
• coding practices, versioning, where code lives?
• storage to use? existing nfs storage? how to stage data?
• Candidate for us to transfer analysis

Next steps

• Answer questions internally (many are sysadmin and policy related)
• Proposal to individual within lab to take over analysis
• Provide details on how to access resources (e.g. HARDAC)