Minus 1 (“-1”) Click Automation – A Metaphor for Cloud & Edge

I chose the title of this blog consciously. It really means “-1” (minus 1) click! Think of it as a metaphor for our Cloud & Edge use-cases.

Let me give an analogy here – Traditionally we have following 2 highly optimized no of clicks paradigms when it comes to solving automation problems – here you can think in context of automation for Cloud or Edge computing use-cases.

1) 1-click: this is automation happening for specific use-case after user gives consent to carry it out. So here manual intervention is in picture to trigger necessary automated workflows. One example of this is our e-shopping cart getting processed to place the order for required items, payment happening through default/preferred method, shipment to preferred address. At the background necessary workflow to warehouse is triggered to initiate the item allocation to desired consumer & shipment.

2) 0-click: this is automation happening for specific use-case without user giving any manual consent or input. So no manual intervention is in picture here to trigger automated workflows. It could be a pre-configured workflow which gets activated upon hitting specific event/threshold to take desired action(s). One example here can be AWS CloudWatch raising specific alarms upon hitting predefined threshold for CPU Utilization of 85% for say 5 minutes. This can invoke a SNS notification to notify required Edge devices accordingly. Note here that this is a “Reactive” stand i.e. Automation happening after specific event has happened. And this is NOT a Proactive stand i.e. automation kicking in proactively before the event.

With this background, now if you think about “-1” click Automation – it’s a Proactive Automation, without any user intervention, for Cloud & Edge use cases. It’s not a type of 0-click because it can have notion of “Intuitive” Automation as well. Interesting!

Let’s step back a bit to see the big picture here. We can define model for resources to be managed/automated across Cloud or Edge (I’m not adding any Fog related elements here for now https://en.wikipedia.org/wiki/Fog_computing – may be in near future). Such model definition is one of the important catalyst for us to achieve necessary automation. Of course it starts with discovery to create a raw model & define a Virtual Construct; but then the Intelligence lies in defining it in such a way that it gives us Autonomy to achieve “-1” Click Automation. Is it possible?

Other interesting fact is about making such model Intuitive to handle obvious scenarios as well as less obvious scenarios. How can we do it?

We are at an interesting point wherein Computing is in the Air J–
Cloud <-> Fog <-> Edge

Network is the Computer by Sun Microsystems is more true now!

Reality with Cloud today is –

  1. a) Hybrid i.e. combination of both On-prem (includes Private Cloud) & Public Cloud i.e. famous 3 – AWS, Azure, GCP (and others) as well as
  1. b) Multi-Clouds

Cloud provides Centralized control whereas Edge gives distributed, localized (think like locality-of-reference, caching paradigm), low latency control. Combination of Cloud + Edge gives us much powerful Engine to handle many relevant use-cases ranging from super critical healthcare system managed by Intelligent Edge & less Intelligent Cloud to relatively less critical system of surveillance having moderate Intelligent Edge & more Intelligent Cloud (Assumption here is that we need aggregation for doing analysis & hence more intelligent Cloud).

 

In all this, way to provide “-1” Click Automation which is Proactive, Intuitive & SLA compliant (covered in the last blog) is necessary but are the existing paradigms sufficient to provide so?

 

Intelligent IOs,

IOPhysics Systems