Helping Cloud Professionals understand the building blocks needed to construct multi-cloud infrastructure

There is a massive on-prem infrastructure build about to take place in the mid to large enterprise as evident in the $600B+ 2021 Venture Capital start-up investment bonanza. Let me explain.   

Application developer requirements are forcing infrastructure teams to build and run multi-cloud infrastructure. That is, application developers have allegiance to cloud service provider (CSP) tools and not necessarily CSP brands. A software engineer will write an application in any CSP or private cloud that is best suited for their application so they can create business value quickly and at scale.  Regulators are also starting to push multi-cloud too as a means to mitigate concentration risk. So to provide developers options and choices of where they build digital corporate value, infrastructure teams are responding en masse by developing architectures that connect, secure, observe and automate application portfolios that can be hosted across multiple clouds (private and public).  Multi-cloud is a massive, trillion dollar market, with much of the build taking place on-prem. On-prem is where multi-cloud happens.

There are multiple definitions of hybrid and multi-cloud. At ONUG we have focused on the term multi-cloud. Hybrid is a term that means infrastructure that spans on-prem and off-prem. There are products such as AWS Outpost, GCP Atheos, Azure Stack, etc. These products extend tools, controls, connectivity, observability services native to the specific CSP on-prem. Multi-cloud is the ability to build, run and manage applications across multiple CSPs and on-prem. Multi-cloud is CSP independent and indifferent. You can’t expect multi-cloud to come from the CSPs; it will come from equipment suppliers and SaaS offerings. This is where the massive build is starting to happen.

Today a multi-cloud strategy is much like enterprise computing in the late 1980s/early 1990s. Each CSP offers unique connectivity, security, observability, application development tools, etc., much like Apple and AppleTalk, Novell and IPX, Digital and DecNet, IBM and SNA, etc., did a few decades ago. That is, each computing company offered unique services addressing different enterprise needs, but from a management point of view, as enterprise computing grew in the PC and LAN era, so did massive management bloat and complexity. It was only a matter of time until complexity stole the value offered by these computing companies. This is about where multi-cloud is today, but only worse. CSPs don’t offer the controls and compliance tools corporations are used to that allow them to mitigate risk and protect their brands.  This was a nonissue back in the PC/LAN era as all computing was on-prem.

Even with this risk to manage, enterprise computing is multi-cloud. CSP application development tools should be rich, unique and expose special CSP features to software engineers. While there may be a standard or two that emerges over time that makes multi-cloud infrastructure easier to build, today, infrastructure engineers have to stitch together their own solutions. In the PC/LAN era, eventually TCP/IP became the standard that the internet was built upon and every corporation adopted, worldwide. This gave developers immense freedom and drove the dotcom period and is now the basis for the digital economy.

In multi-cloud, the good news is that there are so many new companies as well as established tech firms that are offering building blocks for multi-cloud. Companies such as Prosimo, Netris, Aviatrix, Alkira, Pensando, Arrcus, ZPE Systems, Forward Networks, Opengear, NetOne, Augtera, Elisity, NetBrain, Concourse Labs, Wiz, Sysdig, Gluware and so many others are. Many of these companies are only a few years old and are poised to be multibillion dollar entities. Established players such as Cisco, HPE, Broadcom, Zscaler, Oracle, Google, IBM, Microsoft Azure are now all offering and focused on multi-cloud solutions.

At ONUG Spring, many of these companies will be demonstrating how they connect, secure, observe and automate multi-cloud infrastructure. You learn so much more when you interact with technology so we are planning live demonstration pavilions for the ONUG Collaborative Working Groups on Orchestration & Automation (observability) and the Cloud Security Notification Framework (CSNF) canonical data model and decorators. The CSNF demo will highlight the MVP (Minimal Viable Product) iteration of CSNF. Conference attendees will get to see the real value this open source initiative will bring to the enterprise as we normalize and decorate “anomalous behavior” messages. There will also be examples presented by third-party providers, as well as a glimpse of the extensibility of the framework in other domains, like compliance!

The purpose of the live demonstrations is to help IT professionals in the mid to large enterprise understand the available building blocks to construct multi-cloud infrastructure to enable their application developers. At ONUG, you’ll hear from your peers, white board solutions with technology providers and gain confidence to start your multi-cloud journey. When you see these products in live demos, you’ll gain confidence that you can use them, and then you’ll understand why we say “You know it works because you saw it at ONUG.”