2021. What an interesting year. With the world turned upside down by a pandemic that seemingly had its sights set on...
Social Media Round-Up [June 2020]
Keith Townsend develops podcasts and posts and is a member of the HPE Influencer and Blogger groups. Keith is also a frequent contributor to LinkedIn and his articles are widely read (and commented on) by his CTO following. His insights can also be found on twitter.
Keith and I often chat about technology even though our backgrounds are quite different in terms of systems we have worked on. He has a solid following among CTOs and that’s something I cannot ignore. So it was with keen interest that Keith launched his latest endeavor, building a modern “enterprise data center.”
How many times do we hear the response to any promotion of NonStop that NonStop simply costs too much and yet, a basic x86 server equivalent to a pretty basic NonStop system can cost more than the NonStop system. Furthermore, it’s not even a fair comparison as NonStop includes the software stack whereas the x86 server equivalent still has to price in the stack. And much more!
As you can imagine I was most interested in the costs, the technology and the performance of the proposed system. Over the course of a week, I was treated to the following exchange that I think will be an interesting read. It’s all good input and yet another touch point about just how affordable the new NonStop systems have become. Surprised?
My “lab” that I’m building to tell the story of modernizing the enterprise data center is costing roughly $17,666.00/month or $212,000/year. That’s before we modernize the data center by connecting it to the public cloud via a 10Gbps network connection and upgrading the hardware to the latest Icelake Intel platform.
We are doing cutting edge storytelling over here at The CTO Advisor LLC. Make sure to follow the hashtags #CTOAHI to follow our journey. We’ll be provisioning a 10Gbps network connection to connect our legacy DC up to the public cloud thus creating The CTO Advisor Hybrid Infrastructure and telling the story from there.
I am looking at this as I have followed Keith for some time. But here’s my question – what exactly is modern and what must you have before you can consider having a modern solution? What are you prepared to give up in order to be modern … in all my years this is a moving goal line as even today I am being advised that Java in all its flavors is legacy just as Unix dropped off a cliff and it too is legacy – are we prepared for a post cloud, post Linux world?
Those are some really great questions. Honestly, I go back and forward about what’s modern and what’s not. People are still buying IBM Z for greenfield deployments. Is that not modern? It may be too ambitious to tack the topic of defining the ever move the definition of modern. It’s a point-in-time definition. My data center story is at least a 3-year journey. So, what’s modern at the end of this year may be legacy by the end of the life of this project.
Me again, Keith – one item? Modernizing your enterprise data center; what performance numbers are you looking at – number of transactions? number of end users? number of data base accesses? As I look at your costs I don’t have any idea how big a system this is or any idea what it can achieve performance wise … anything you can share?
Wow… I love this question. Were you sitting in our team meeting yesterday? Since we couldn’t find an anchor sponsor like SAP to sponsor the workload, we have to identify some target use case and metric. We are developing the requirements of that part of the story this afternoon. Because, what says our DC isn’t modern enough? The why modernize is just as important as the what we are modernizing. Part of it will be driven by system performance. Another part of it will be driven by operational agility. We are defining the use case and metrics and will share them as part of the story arch.
I can count on Richard Buckle asking the most important question about The CTO Advisor LLC Hybrid Infrastructure #CTOAHI) modernization project. Great, I’m spending $212K annually on this project. Cool number. But why? Not why am I telling the story or even building the data center? Why am I upgrading/modernizing the data center? What performance or business indicator am I using to determine I need to move from a relatively new HCI stack to something different.
We are fleshing out the use case and scenario. As parts of editorial and sponsored content, we’ll tell what are some of the advantages of adding an Oracle Cloud VMware Solution to our stack vs. simply doing an in-place upgrade.
Thanks to Richard for the setup!
I wait with bated breath …
It was also Keith who provided a fresh insight into HPE in a video he posted during the HPE Discover Virtual Experience. I won’t add anything further at this time other than to suggest giving Keith five or six minutes of your time to view his perspective on where he thinks HPE is headed and why he is now bullish on HPE as a whole going so far as to suggest that HPE now possesses superior products:
As an aside, this is setting up to be the perfect introduction to a series of posts on NonStop that revisit the topic of TCO and of the affordability of NonStop. It will be good to get back to Keith and compare what he is doing with what NonStop provides out of the box.
And on that note, we can finish with a reference to the latest YouTube video by our own Steve Kubick. As for topic you should be intrigued by his focus on trends and on fashion and on his dad’s admonition that what once was new comes around every generation. Yes, the title My Dad’s Paisley Tie and Cloud technology should no doubt pique your interest!