Global Airline Outage – One Vendor Chaos

swahuffpostoutage

Vendor and supplier management is amongst the most boring topics to explain to your spouse, in-laws, and friends when they ask “What exactly is it that you DO for a living?”

It is always great to have a real life example to help. Today’s apparent failure causing chaos nationwide is PERFECT. Many birds, one stone.

A company, Amadeus, has successfully innovated and created a platform for airline operations that many airlines have adopted. Not surprising actually since user friendly and airlines have rarely been expressed in the same sentence.

Apparently this morning there was a problem. A large one. One that affected most/all airlines using the system. Reservations, kiosks, gate agent operations, et al.

In June 2016 I snapped a photo of Amadeus in action at AUS. It was in stark contrast to the text screens that agents used in decades past. Note the ability for multiple airlines to share one gate. Awesome. If it works. ALL the time.

Questions

  • Disaster Recovery? – Amadeus? Airlines? Why are passengers affected when a single vendor fails?
  • Regulators – Is it acceptable for a single vendor to be allowed to substantively disrupt airport operations globally?
  • Attackers – Want to inflict max chaos and disruption to critical infrastructure? A single vendor may all you need to target.
  • Protectors – Do you trust your vendors? Based on what evidence?

Early Adopter Pain and Hyperconverged Excellence

Austin’s @pivot3 recent tweet of an @ITWire article about the projected growth in the HCIS (Hyperconverged Integrated Systems) market contain a couple of nuggets of information that demand expanding upon.  

First:  Hyperconvergence has yet to enter the mainstream.  

This is true.  The article states that while HCIS is the fastest growing segment it will be less than 1/4 of the integrated systems market by 2019. 

Innovators and early adopters are clearly benefiting from the fruits of hyperconvergence, namely software centric infrastructure built upon commodity hardware to deliver transparent solutions to internal and external stakeholders quickly and efficiently. That said early adopters are also on point to absorb the inevitable pain of industry upstarts and established giants finding and exploiting a particular market niche and redefining their solutions on a path to revenue nirvana. 

Pivots, acquisitions and chaos lay ahead for some. As evidence some friction has arisen between Dell and EMC as they struggle to understand how Nutanix fits into their new world order. 

Secondly:  Hyperconvergence use cases have been limited and silos have emerged.  

No kidding.  If you treat hyperconvergence like a traditional enterprise point solution, roll with a limited POC, and get frustrated when the touted benefits are not realized, then you deserve your fate.

Hyperconverged excellence requires a philosophical shift to a software first/software only mentality from the help desk to the CIO.  Hyperconverged excellence requires that you are willing to shrewdly assess your organization and act to make the necessary changes to be successful.  Blockers must be eliminated be they technological, cultural or personnel. Hyperconvergence holds great reward only for those organizations that understand the revolution at hand and have the courage to execute.  

The irony is that those that languish in perpetual POC hell will eventually abandon their infrastructure entirely for a public/private cloud solution managed by someone that achieved hyperconverged excellence long ago.

Austin’s Loop 360 Corridor is a terrible place (for your business)

Austin’s Loop 360 (Capital of Texas Highway) long ago envisioned as a loop around the west side of Austin, but now consumed by growth, is home to many life draining fluorescent prisons, more commonly known as “Class A” office space.  Several high profile technology companies have inexplicably decided that it would be a good idea to locate their workforce along this stretch of roadway.
Anyone who spends any amount of time in Austin understands that traffic is undoubtably the biggest negative of living or working in or around Austin (hipster culture is a close second).  Traffic alone though is not what makes this particular “highway” a terrible place.

The problem?  Any network architect would have ample time to tell  you as they waste precious cycles sitting in traffic that could be spent doing something productive…  SPF or “Single Point of Failure”.

Many offices along 360 have only one way in and one way out.  No alternate routes unless you have four wheel drive and a chainsaw…  Any incident along this single path makes an already ridiculous commute, untenable.  A serious incident like a wildfire or intentional act by nefarious actors could turn inconvenience into disaster.

My recommendation:  Always consider ingress and egress under normal and emergency conditions when evaluating any location which will contain your most important resource; your people.