19 lines
1.2 KiB
Markdown
19 lines
1.2 KiB
Markdown
## Databricks Incident Analysis
|
|
|
|
Refer to https://git.kalar.codes/NickKalar/Incident_Report_Generator for more context.
|
|
|
|
After using pure Python and Pandas to create an Excel Spreadsheet to compute analytics for a CSV
|
|
file of Support Incidents, I decided that the application I built was too cumbersome and not easily
|
|
extendable. It would require a lot of fine tuning, and trial and error to do so much as add a new
|
|
table or graph. Using Databricks, most if not all of the same analytics can be created in a simple
|
|
and extensible way so that even the least technically savvy can add or manipulate data as needed.
|
|
|
|
## How to use
|
|
In Databricks, create a catalog named `analytics`, inside that create a schema named `incidents`,
|
|
and inside that create a volume named `current`. Upload the CSV file here, or create your own with
|
|
the same headers as the file here.
|
|
|
|
Once completed, import the notebook and run the first cell to get the majority of the analytical
|
|
work down. If the source system can be scheduled to dump a new CSV of incidents into an S3 or
|
|
similar storage system, this could be updated to run on the same schedule (offset by a few minutes
|
|
or more) and alert stakeholders on success or the developer/admin on failure. |