When I hear the term "power trio", I think of Geddy, Neil and Alex.  A band, where there's just a bass, drums and guitar, with one of  'em singing.  By the way, if you like Rush, I'm guessing you'll like Winery Dogs too.  Both are outstanding examples of power trios.  Another great example, in, umm,  a very different genre, was unleashed on the world earlier this spring.  That's when Cisco announced a power trio that makes our overall Data Center solutions sound even better than before. CliQr, HyperFlex and new Nexus 9000 switches with Cloud Scale technology make hybrid cloud better.  These allow customers to optimize and integrate private clouds running in their own data centers with public clouds, leveraging policy to enable automation.

Why did we introduce these new components?  And was that too rapid a turn from the musical metaphor?  (I suspect the response is contingent on your musical taste.  In any case, I'll only address the former question below.)

• Hybrid cloud will be the norm for most enterprises.
• Integrating all the pieces and moving apps across multiple clouds can be complicated.

Most customers currently have their own data center(s).  They're migrating to private cloud, so they can respond more rapidly to business requirements, etc.   They're also using public cloud, and often public clouds (plural).  So, by definition, they are moving toward hybrid cloud environments.  ZK Research estimates that 84% of organizations plan to deploy hybrid clouds.

So, given these thoughts, a couple questions:

• What happens when public cloud provider A raises pricing and you want to move your workloads from provider A to provider B?
• What about provider C, who provides SLA's at a level nobody else can match, that are absolutely necessary for a subset of your critical apps?
• How 'bout your own private cloud, that is now tricked out to the point that you can deploy apps as fast as any external provider, at a lower overall cost, with higher security?
• Maybe you just wanna move from dev on provider A to production in house?

Changes aren't permanent, but change is.  The point is, workloads can/should be able to move to/from different clouds for different reasons over time.  But doing this can prove to be quite complicated for a variety of reasons.  Typically, this can involve cloud specific scripting and image formats, the need for code modification, yadda yadda.

In any case, in an optimal scenario,  it would be good to:

• Get to a point that your private cloud is 'all that'...So the local data center network can handle things such as the imminent explosion of endpoints resulting from containers and the micro services that reside on top of them.
• Leverage a hyperconvergence solution to do rapid rollout of storage and compute with simplified operations.
• Move workloads from cloud to cloud with the lowest amount of time, hassle and cost.

Key in doing much of this is the notion of a generalizable profile, that expresses the requirements for the application in a way that is portable across different clouds, and to have it work in tandem with policies that automate as much as possible.  This should sound familiar...UCS...ACI...there's a pattern emerging here with this whole policy based profile thing.

In any case, if these topics interest you, please check out this report from ZK Research that investigates Cisco's power trio and read how it makes hybrid cloud sound better than ever.  Also, check the links up top if you want to hear a couple other great power trios.  Go ahead.  Click the links now.  If you choose not to decide, you still have made a choice.

(Note: Italicized phrasesare Rush lyrics.  But you probably already knew that.)

If you get a chance, please reply because I'm interested in hearing what you think about any of the links!