Build or Buy Cloud Infrastructure?

by Nick Lippis

On August 17th, Cisco reported that revenue fell for a seventh straight quarter. IBM has experienced revenue decline for the past 21 quarters. HP has been broken into multiple businesses. Dell was brought private after it acquired EMC and VMware. All the while, Amazon, Google, and Microsoft’s cloud businesses have grown stronger and collectively invested $31.5 billion in 2016 on capital expenses and leases, up 22% from 2015. Bridges are being built to extend enterprise private cloud infrastructure to public thanks to a new crop of software-defined startups. When you start connecting the dots, what you see is an IT industry that is fundamentally changing. Corporate IT planners always speak with their budgets, which represent their ultimate power, and ever since ONUG began in 2012, those voices are speaking with increasing volume.

Driven by the business imperative of digital transformation and illustrated through recent collective purchasing trends, we can see that IT executives are investing in multi-cloud software-defined infrastructure products and services. To increase agility, seek favorable economics, and design flexibility, they are moving away from a hardware-based model of infrastructure to a more nimble software-defined model.

In this new Application Programmable Interfaces or API-driven world there are new demarcations, increased design flexibility, and product “extensibility.” That is, the ability to extend a product’s feature set or value through programming via APIs. In fact, we have entered an API-driven era where IT organizations are stitching together solutions that extend their on-prem cloud infrastructure to off-prem as well as splitting workload between the two.

Consider this: during ONUG Spring, 71% of the ONUG IT Community said that within three years they will no longer build private data centers!

With the cost of a new data center reaching approximately $200 million, we asked, “how will that IT budget spend be consumed?” Approximately 77% of the ONUG IT community said that they would shift between 10-50% of their IT infrastructure spending to the public cloud.

If this comes to fruition, which ONUG highly anticipates, then this is a massive redistribution of spending, a new approach to infrastructure design, new vendors and relationships. By “vendors,” we mean IT vendors, as well as cloud and service providers. ONUG tracks the buying power of its community, which is larger than $100B in annual IT spend. If 77% of that spend moves to the cloud over the next three years, then this is indeed a massive shift in consumption models.

One of the biggest changes occurring at the IT executive level is in the way these IT business leaders think of their IT vendors. They used to rely upon them in their respective silos. If there were a problem, the IT vendor would fix it. If the IT business leader were deploying a new workload, all vendors in the workload’s dependency map would collaborate to support said workload. In short, IT business leaders relied upon the relationship with their IT vendors for innovation, support, and service.

This is changing at a fundamental level. IT business leaders are increasingly viewing vendors as modules or pieces of a fabric that are stitched together via APIs, where control and innovation are shifting to IT departments with programming expertise and rooted in a new enterprise-based DevOps culture. For example, an enterprise monitoring solution today consists of a group of vendors, each providing a specific expertise, be it data collection, correlation, data ingest, analytics, analytic engines, etc. All of these products are integrated by the IT department or their consultants via API to deliver the designed monitoring outcome, perhaps automated alarm event mitigation, for example. The new thinking is that IT business leaders are relying upon their IT teams to envision, design, and deploy solutions via APIs that extend each product into a hybrid multi-cloud fabric that is increasingly software-defined.

The IT world used to work by people/companies banding together to create standards and/or open source projects. Enterprises relied upon those standards and/or conformance with an OS project to guide their purchases, even if what they bought was just a support contract. Will enterprises continue to consume products that conform to standards and/or mainline open-source distributions, or are we turning a corner where API’s have lowered the bar of entry to the point that enterprises will increasingly build their own infrastructure software/hardware?

The large enterprise market seeks instantiation of on-demand capacity and agility, which is what the cloud providers are delivering. However, this is bracketed by economics where the cost to a large data center is approximately $200M of capital outlay and favorable cloud economics break down at approximately 15K virtual machines. Further, with the right architects and engineering leadership, large enterprise IT departments can build IT at lower costs than cloud providers since they don’t have to be concerned with profit margins. The issue is that not many enterprises have the technical leadership.

The question of whether to build versus buy cloud infrastructure will be a main theme at ONUG Fall in NYC on Oct 17th and 18th hosted by GE. Ernest Lefner of Bank of America will connect the dots as he moderates a discussion with Tsvi Gal (Morgan Stanley), Chris Drumgoole (GE), and Swamy Kocherlakota (S&P Global) to see what kind of picture emerges.

 


Author Bio

Nick Lippis pictureNick Lippis

ONUG

Nick Lippis is an authority on corporate computer networking. He has designed some for the largest computer networks in the world. He has advised many Global 2000 firms on network strategy, architecture, equipment, services and implementation including Hughes Aerospace, Barclays Bank, Kaiser Permanente, Eastman Kodak Company, Federal Deposit Insurance Corporation (FDIC), Liberty Mutual, Schering-Plough, Sprint, WorldCom, Cisco Systems, Nortel Networks and a wide range of other equipment suppliers and service providers.

Mr. Lippis is uniquely positioned to comment, analyze and observe computer networking industry trends and developments. At Lippis Enterprises, Inc., Nick works with entrepreneurs evaluating new business opportunities in enterprise networking and serves as an independent investor and advisor.

 

 

 

Author's Bio

Nick Lippis

Co-Founder and Co-Chairman at ONUG

Nick Lippis is an authority on corporate computer networking. He has designed some for the largest computer networks in the world. He has advised many Global 2000 firms on network strategy, architecture, equipment, services and implementation including Hughes Aerospace, Barclays Bank, Kaiser Permanente, Eastman Kodak Company, Federal Deposit Insurance Corporation (FDIC), Liberty Mutual, Schering-Plough, Sprint, WorldCom, Cisco Systems, Nortel Networks and a wide range of other equipment suppliers and service providers.

Mr. Lippis is uniquely positioned to comment, analyze and observe computer networking industry trends and developments. At Lippis Enterprises, Inc., Nick works with entrepreneurs evaluating new business opportunities in enterprise networking and serves as an independent investor and advisor.