Predictive analytics to the server rescue

role in the project
Lead Product Design
duration
10 Months
Date
2015
Project
Nexenta
Nexenta - Data Center Storage Management

Keeping up with the up-times

Our digital lives are expanding faster than the universe. Your banking information, family photos even your pet records are in the cloud, and the cloud is in a building in some data center. All your stuff resides in a temperature-controlled, buildings with massive rooms stacked full of servers. Servers like any other machine, periodically breakdown and need to be fixed or replaced.

Predictability

If you know that a server breaks after 8 trillion actions involving storing and retrieving data, then you should probably buy a new one before reaching that milestone. Otherwise, you and your clients will be dealing with costly down-time. Similarly, if you know that for every million users you need to buy more data space, then you can probably budget your server expenses based on the growth of your user base.

Defining the problem space

I had an amazing opportunity to work with folks at Nexenta to help build a product that enables them to manage servers and monitor future growth more effectively. 

Working with people that set up servers and maintain data centers was really interesting. I learned quite a bit about how servers work and the challenges they face. Most of this work occurs at a terminal, which requires knowing the programing language. Building a WYSIWYG dashboard, which is what I was involved in, provided a huge advantage for people who may know servers, but don't know how to use a terminal.


Login Screen
Dashboard
Visual representation of a server
Visual representation of server structure
Showing a state of "in-progress" of a server setup that could take a few minutes.
Custom report builder
Setting up preferences, the most interesting part about this design was having to learn all these terms about servers.
This is where you'd start setting up your cloud (if you had one).