Apstra has a webinar that helps answer a lot of questions around how we are able to dramatically decrease time to market for new data center deployments. As part of the webinar, I will be providing a scenario and showing how AOS® 2.0 automates the design and operations of complex networks and server compute pools. Here is a sneak peak of what we will cover.
Suppose your CTO has recently told you that the Finance department has decided to create a new application to handle an upcoming new product launch. You have heard a bit about the app and know roughly what to expect, and you’ve built similar environments in the past. Other architects in your organization have promised the CTO it will take two months to create and test the new system before handing it over to Finance for the application to go live.
You’ve built a similar system using VMware ESX and your standard Arista and Cumulus hardware switches, so you know what it will take to complete. You have to create a new diagram, cabling cutsheets, device configs, and once the system comes online you need to get all of the necessary elements into your common monitoring platform so the NOC can take over the daily operations. You have the hardware and need to complete the cabling, which might take a day to patch correctly. All in all you are thinking this will take about two weeks to finish, which is a big improvement over the two month time window that the CTO expects.
But what if you could provide some basic logic, or what we call “Intent”, and have the system configuration completed in 15 minutes? Once you have the configs on the devices you can start patching the cables in and testing everything. Then comes the inevitable challenge of finding the one or two typos you made in the configs or cabling matrix, which always takes too much time. Finally you need to get all of the elements including routing, switching, physical connectivity, and performance monitors into the monitoring system.
Would you get recognized for incredible service if you could do this all in one day? Would the CTO ask you how you were able to beat their expectations by many weeks, and would this result in improved chances for a raise or promotion in the future? And would you be able to sleep at night knowing that the network was built perfectly with no mistakes that might have the NOC waking you up in the middle of the night when they found the one config error you had missed?
We feel that the answers to these questions would be an undoubtable “Yes”, and on this upcoming webinar we’ll show you how AOS can do exactly what we just described.