In this article, learn about the transition from FireCloud to Terra, and some favorite features, mighty or small, that didn’t exist in FireCloud but revolutionize getting things done in Terra.
In case you haven't met us yet, Beri and I (Sushma) are part of the Frontline Support Team here, aka the real humans behind the FireCloud and Terra forums and helpdesk. What you may not know is that we use FireCloud and Terra to create and maintain many of the Featured Workspaces, which on any given day might involve managing data, writing WDL workflows or hammering out some R or Python code in a Jupyter Notebook.
A much improved data management interface
- We hate having to deal with the FireCloud API to delete rows in the workspace data table. Terra has blessed us with the ability to delete data entries by simply checking a box and clicking on that sweet Trash icon.
- Also a huge annoyance in FireCloud, the rigid data model that forces you to mush your data into the cookie cutter TCGA schema of participant, sample, pair and so on. In Terra, you can just upload any custom table you like. Ah, the freedom.
- We also really like the new File Browser in the Data tab - you can browse your bucket data without having to wrangle the Google Console in a separate browser window. Incidentally, you can select files from here to add to your workflow configurations in the wools tab too.
For those of you who are new to Data Model shenanigans, trust us when we say these features are pure gold.
A preview mode for notebooks
The engineers say there's not much they can do about the time it takes to spin up the notebook virtual machine (VM) (or cluster), as that's apparently a Google thing. But they did put in an awesome preview feature for notebooks that lets you read through the contents and refresh your memory while you wait, which is quite an upgrade from watching a loading cursor.
It's way easier to retry failed workflows
Shockingly, there are times when my workflows don’t run perfectly. In FireCloud, this means I have to wrangle the list of failed workflows, create a brand new submission, and then start all over. On a large dataset, this gets beyond groan-worthy really fast. In Terra, with the click of a button(!!) I can retry just the failed workflows. So simple, yet so useful… There's a definite "one-click to joy" theme emerging with these Terra feature updates; we're loving it.
Overall it's nice that the newer greener Terra is easy on the eyes, but really, it's what's on the inside that matters, and we have truly got all the feels for these super helpful new features. Check out the app and let us know what are your personal favorites in the comments!