When To Cloud

Is cloud access and analysis for you?

Cloud migration can often have a steep learning curve and feel overwhelming. There are times when Cloud is effective and times when the download model may still be more appropriate. Here we aim to help people decide what’s best for their use cases.

Questions to Ask Yourself

• what does it mean to be in the cloud?
  • accessing services via internet? no
  • logging into an AWS/Azure/etc account (awareness)? yes
    
• include workflow type figures and discuss
  • link to any relevant tutorials that demo these workflows
  • how can you “cloud”?
  • what’s already available (services that assist)
• listing questions from Andy’s slides
  • a bit of discussion + examples for each question
  • concrete examples accomapnying each question
  • “I have a TB of storage on my laptop….” storage scenarios
  • see this from Ryan Abernathy: https://medium.com/pangeo/closed-platforms-vs-open-architectures-for-cloud-native-earth-system-analytics-1ad88708ebb6
  • include a question on actions that are already services (subsetting, etc.)

Challenges

It is important to be aware of the drawbacks and challenges associated with working in the cloud. Common feedback from early cloud adopters are summarized here:

• feedback from Champions/workshop participants
• (see Alexis ESIP poster)
• (this could be its own section on website?)

Considerations

We are now accustomed to living in a highly digital world, separated from physical reminders of the services we use. No longer do we access documents from a row of filing cabinets, we now store them in cloud-based archives (e.g. Google Docs). We run code on high-performance computing clusters, removed from the whirring and warmth generated by servers that live away from our desks. The following are considerations for using cloud-based resources:

• separation from physical signs of compute power (using resources/energy)
• physical location of “the cloud”, whose land is it on?
• testing/playing locally before migrating workflow to the cloud
• working remotely (not WFH, but computer is remote)
• environmental impacts