Project

General

Profile

UseCases » History » Revision 15

Revision 14 (Sarah Zaranek, 09/20/2022 03:21 PM) → Revision 15/27 (Sarah Zaranek, 09/20/2022 03:22 PM)

h1. Use Cases for Workbench 

 h2. Running/Testing Workflow that I wrote  

 I want to run a workflow on a collection that exists 
 * did my workflow run correctly? 
 * if finished - check status, logs and output 

 h2. Running a Workflow that I didn't write  

 * Assumption that workflow already works. 

 h2. Sharing Workflow I wrote for someone else want to run 

 I want to a person I work with to run this workflow I wrote on their data 

 that someone built for me.  

 h2. Debugging Workflows 

 I have a collection with buggy/suspect results. I need to figure out where in the pipeline the bug crept in. 
 * any big differences between the different "foo_XX" steps? 
 * flags passed to the "foo_XX" steps 
 * logs from one (all?) of the "bar_XX" steps 
 * docker image metadata (docker image ID, name, version, dockerfile) 

 h2. Finding    and Comparing Workflows  

 I need to find a "good" workflow invocation, and compare it to something that isn't working now 
 * when did this workflow last succeed? 

 h2. Creating a collection 

 I want to create a new collection 
 * from data already in an existing collection 
 * from a subset of files in an existing collection 
 * combining files from different subsets of different collections 
 * uploading from my desktop 
 * downloading from ftp, s3 or other location 
 * from the results of a already run workflow 

 h2. Getting Collection Lineage 

 I have a collection that is valuable/interesting. I need to review/share/publish how it was generated (tools/versions/options/inputs). 

 h2. Find a Collection 

 I need to find some data, and confirm I'm looking at the right collection.  

 h2. Cost Estimation 

 I want to estimate how much money it will cost to run a workflow. 

 h2. Run-time Estimation 

 I want to know how long it will take to run my workflow. 
 * queued? when will it start? 
 * taking a long time? how long does it normally take? 
 * slower than it should be? should I run it with more CPUs, RAM, etc.?