Today’s Datanauts episode was recorded live on the showcase floor at VMworld 2016 in Las Vegas. Chris Wahl and Ethan Banks opine about what engineers are supposed to do in the changing world of enterprise IT.
They talk about emerging data center themes including automation, containers, private and hybrid clouds, the rise of open source, and more.
They also look at the implications of these trends for professional development and career advancement. Where should you direct your energies? Should you get certified? Do you need to learn Linux?
Joining the Datanauts to share his own opinions is Justin Warren, founder of the research, advisory and consulting firm Pivot Nine. He blogs and hosts a podcast at EigenMagic, and you can follow him on Twitter.
This episode of Datanauts is brought to you by ITProTV. Enhance your technology aptitude. ITProTVis the resource to keep your I.T. skills up to date, with engaging and informative video tutorials. For a free 7-day trial and 30% off the life of your account, go to itpro.tv/datanauts and use the code DATANAUTS30.
Part 1 – Themes We See In Infrastructure
- GIFEE (Here’s the Datanauts podcast about it.)
- Hybrid cloud
- Not all is going to public cloud.
- Privacy, performance, control are all big issues for the enterprise. We’re not outsourcing everything to public cloud, and some stuff that was in the cloud is coming back.
- Yes, you can do containers on Windows.
- BUT, note that MS is getting closer to Linux each day.
- For example, PowerShell is now available on Linux.
- Automate all the things.
Part 2 – What All This Means For The Real World
- You don’t have Google problems.
- But maybe automation isn’t a bad thing.
- Reduction in error rate
- Abstraction doesn’t make infrastructure less complex. It makes it more.
- Infrastructure engineers will need to know more, not less.
- When it works, it will be amazing. When it doesn’t, it will be hard to fix.
- Companies outsourcing IT responsibility to vendors (hyperconverence) and to XaaS providers.
- Could that mean some infrastructure engineers aren’t required? (i.e. replaced by software)
- The silos are coming down.
- Can’t cling to esoteric knowledge as a foundation of value.
- Know more about more.
- Play well with others.
- Speak business.
Part 3 – What To Bet On In The Future
- Deep vendor plays – this is where the ultimate integrations are really happening for enterprises. In unified stacks provided by vendors.
- Telcos are rolling their own stacks, but they are atypical — not previews of enterprise networks to come. Different problems, different solutions.
- But there is overlap.
- Agility – easily stand up an application. Call that devops if you like.
- Make the most of resources – don’t waste CPU and RAM.
- Vendors that can deliver unified stacks.
- VMware. They are territorial. But NSX solves network segmentation issues.
- Cisco, but they don’t have a strong storage answer, and still experience fighting among BUs. Too many solutions, not a centralized vision.
- HPE. All the pieces, but have gone through repeated re-orgs. No evidence yet that HPE has sorted out how to sell a unified platform.
- Dell. They have all the things now. EMC, VMware. Lots of servers, storage, networking. Can they put all the pieces together and convince the enterprise?
- And what of VCE?
- The “less big” hyperconverged players, including Nutanix.
- Open source projects gaining momentum.
- Kubernetes – scheduling software for containers.
- OpenStack – um…where does this go long-term? Arguments being made that Kubernetes + containers = OpenStack replacement.
- Public clouds
- AWS is the strong leader. Pay attention to serverless as well.
- Google Cloud is immature.
- The Azure value prop continues to grow.
- Summary of skills that will be valuable.
- Private cloud. Unique skill.
- Not to diminish VMs, but some suggest stateful containers are next.
- Public cloud. Unique skill
- AWS, if you have to pick one.
- Hybrid cloud. This is a scheduling problem. Where do workloads stand up? Public or private? And what’s the networking look like between them? And what’s the required application architecture to make this work? Cisco can play here. And VMware. And others. Look at market momentum.
- Take special note of RedHat. They are craftily in the middle of everything in infrastructure.
- What are your customers using? Vendor relationships in the enterprise are incredibly hard to displace. So don’t assume radical change. Assume incremental and/or same vendor + new product strategies from that vendor from enterprises.
- Private cloud. Unique skill.
- Don’t feel like you have to get certified. That’s expensive, and maybe you don’t have the cash. But be able to demonstrate your expertise.