Chuck's Blog: Towards A Unified View of "Software Defined"

ATN-300InsightaaS: Chuck Hollis is a former EMC executive who is now chief strategist for VMware’s SAS (storage and application services) business unit, and author of the well-regarded Chuck’s Blog. In May, Hollis published an interesting post on software-defined storage (featured here on June 2)  - and this month, he has expanded the scope of the discussion to examine the benefits of "a singular, consistent and integrated view of software-defined disciplines." An examination of the steps leading to server virtualization leads to a discussion of networking, and how the process that drove the success of software-defined servers ("a very tactical need for efficiency resulted in a very strategic outcome") is now playing out in environments that are looking to implement micro-segmentation, or the ability to isolate networks behind the network perimeter to harden them against attacks. From here, he returns to software-defined storage, touting its benefits in policy-defined management. 

The important leap in Hollis's post follows from these three discrete markets. He points out that "in each case (compute, network, storage) services are dynamically composed in response to an application’s specific policy." This isn't in and of itself (as Hollis points out) sufficient reason for integrating areas that are typically considered independently - but Hollis makes a sound case for seeing advantages in dependencies, optimizations and consistency. He provides examples of reasonably-common scenarios - for example, analyzing network log data, improving the performance of remote replication, responding to a spike in read requests - that invoke requirements for multiple, related services. There are practical obstacles to real-world integration across domains (such as the need for managers in each to make the decision that optimizes compute, network and/or storage performance and price/value), but Hollis believes that the architectural value of converged software infrastructures - as with converged hardware infrastructures - will attract a substantial following.

In one corner of our industry, we have a familiar discussion regarding hypervisors, or – more precisely – software-defined compute.

In another corner, we have a vigorous debate around software-defined networking.

And, closer to home, a completely separate debate around software-defined storage.

Shouldn’t they all be aspects of the same discussion?  

What do we gain by preferring a singular, consistent and integrated view of software-defined disciplines – and what do we lose by considering them individually, using a traditional lens?

Let’s Start With Compute …

Wrapping your head around software-defined anything can take some serious effort, if my personal experience is any guide.  I do what I can to help explain the key ideas, and why they are important.

A good starting point for wading into the deep end is our familiar server virtualization – something we all have experience with.   Seen through a "software defined" lens, we could better describe it as using application policy to dynamically compose compute services....

Read the entire post: