(1) Mainstreaming public cloud
The year began with fresh evidence of enterprises treating public cloud services as legitimate alternatives to private infrastructure, a view that was only reinforced later on by events, whether it was record attendance at AWS re:Invent, VMware and AWS expanding a partnership that enables legacy systems to run on AWS infrastructure or Oracle making a last-ditch push to establish cloud bona fides and forestall a mass migration from its applications. Nevertheless, we’re still in the early stages of enterprise cloud adoption. As I wrote at the time,
Although the survey illustrates how cloud services have become a viable and increasingly popular platform for enterprise applications, they are competing with decades of enterprise inertia and sunk costs in facilities, equipment and software. Indeed, one estimate pegs the market for enterprise software at almost $400 billion this year, an order of magnitude higher than the expected revenues of AWS and Azure combined. This leaves plenty of upside for cloud vendors, but it also means that we are many years away – if ever – from public clouds completely displacing a majority of private IT infrastructure and services. The state of the cloud in 2018 is healthy and robust.
Amidst the excitement and innovation around native cloud services from the big three, it’s easy to lose sight of the fact that some of the most significant enterprise cloud technology in 2018 focuses on bridging the old and new worlds. For example -
(2) Virtustream is Dell’s secret weapon for cloud-ifying legacy enterprise applications
Dell inherited many valuable technology assets when it merged with EMC, but none is more intriguing than Virtustream, the cloud platform that's the anti-AWS, designed to run large, monolithic enterprise systems on an efficient, scalable cloud environment. One conversation with Virtustream’s co-founder makes clear that what initially seemed like just another cloud wannabe is a very different animal, squarely designed for workloads that aren’t cloud friendly. As I detailed in an article after Dell World
[According to Rodgers] ‘We took a different track and made it our mission to become the leader in the market for running mission-critical applications in the cloud. While AWS focused on serving scale-out cloud native applications, we architected our cloud platform to solve a different engineering problem: to run scale-up, I/O-centric enterprise applications that require higher availability and security.’
I fell into the trap that snares most technology analysts: assuming that because a technology is capable of something that it will inevitably be used for that thing. In the case of Virtustream, I saw an elegant multi-tenant server and storage virtualization stack that was more efficient and reliable than what AWS then offered and assumed that it would ultimately take on AWS in the on-demand, pay-by-the-hour IaaS market. “I missed the fact that Virtustream was overkill for most AWS uses (particularly in those early days of cloud adoption) like test/dev systems, Web or mobile apps with cloud-native backends and DR backup systems. Instead, Rogers and his team saw an empty racetrack, i.e. an untapped market for migrating traditional, mission-critical enterprise applications to a managed cloud service.
Most critical enterprise applications aren't designed to use cloud infrastructure and require some, and in many cases significant, modifications to work correctly on platforms like AWS or Google Cloud. However, as organizations look to exploit the efficiency, scalability and redundancy of cloud services, there’s an ample market in providing a cloud home for these systems of record that are inherently difficult to change. Virtustream has the technology and single-minded focus on such systems to make it a long-term asset in Dell’s portfolio.
(3) VMware fills gaps in cloud integration with sophisticated network and management tools
As enterprises adopt cloud infrastructure, whether to host new applications or offload workloads from obsolescent infrastructure to hosted services like VMware Cloud on AWS, they need to extend existing on-premise networks, data sources and security policies into the cloud. Indeed, surveys show that most organizations are pursuing such hybrid designs, however, they quickly face the many complexities of stitching together environments with very different designs and capabilities.
Hybrid networking and security is an area of acute pain and critical importance and while cloud providers offer some increasingly sophisticated features like AWS Transit Gateway, Azure Security Center and federated IAM services, they fall far short of delivering a turnkey hybrid network, system and security management system. These gaps are being filled by a variety of third-party products, none more complete than VMware's vRealize Suite and NSX network virtualization products. I called NSX VMware's secret weapon, writing:
VMware is building a multi-cloud infrastructure abstraction layer that insulates users and applications from the idiosyncrasies of the various services...The glue that binds these cloud services, enterprise data centers and branch office devices into a cohesive system is networking, in particular, a growing portfolio built atop VMware’s NSX virtual network technology. "NSX was initially seen as a means of managing the virtual networks of enterprise VMs and connecting their virtual switches into a physical data center network, however, the technology’s capabilities and VMware’s ambitions have dramatically grown over the past few years.
VMware executives have cited NSX as a key growth engine and I posit that it could evolve into one of the company's largest products. Caveats include the glacial pace of change within enterprise networking, many unknowns around the evolution of enterprise network designs, particularly the potential decentralization of some infrastructure, aka edge computing and growing competition from other suppliers cloud-agnostic network fabrics like Aviatrix and various container-focused technologies like Contrail, Weave and Contiv.
(4) Oracle pushes cloud infrastructure, but its future lies further up the stack
Virtustream isn't the only company building cloud infrastructure for legacy infrastructure. As I detailed during Oracle OpenWorld, the database behemoth understands the profound threat to its business posed by cloud-native services, but it also knows that fork-lift migrations of Oracle-based applications to AWS or Azure aren't a viable option, leaving companies with expensive modification efforts to make things work. Thus, even as Larry Ellison and Andy Jassy trade pot-shots, the two services continue to diverge in their strategies and infrastructure designs, with Oracle focusing on making a cloud home that's safe for legacy on-premise applications. As I wrote:
Like most established software companies, Oracle is in the midst of a strategic pivot from sole reliance on licensing and service revenue for on-premise deployments to subscriptions for cloud infrastructure and application services. Unlike AWS and Google Cloud, Oracle has focused on building an environment to accommodate legacy workloads and that minimizes the disruption in migrating database-backed applications from on-premise to cloud deployments. As such, its Cloud 2.0 looks more like Cloud 0.5: a great stepping stone from enterprise data centers to cloud managed services.
Indeed, I think Oracle's preoccupation with AWS and the IaaS market is a mistake. As I detailed in a follow-up article, its cloud strategy should instead center on SaaS:
[Oracle's] capabilities aren’t notably superior to those of the big-three IaaS/PaaS providers, AWS, Azure and Google Cloud. Assuming Mark Hurd’s statement that '15 percent of the U.S. corporate-owned data centers shut down,' is true, and it sounds plausible, Oracle has many opportunities to gain IaaS business. Nevertheless, I think Oracle will continue to play second fiddle to Amazon and Microsoft in the IaaS market. "I believe that SaaS is the bigger cloud opportunity for Oracle as customers like Birmingham City University reason that if they are going to move to the cloud, why take half steps by continuing to operate their applications? Instead, they will see the wisdom of allowing Oracle to manage, secure and update the infrastructure and applications. Indeed, given Oracle’s unparalleled expertise in its own applications, it is highly likely to do a better job than the customers could do themselves.
Oracle's strength is in software development, not infrastructure and the company would do better by pivoting to an infrastructure agnostic approach akin to SAP or Adobe while aggressively expanding its SaaS portfolio. Such a strategy would allow Oracle to exploit managed infrastructure from regional providers that could improve performance in particular geographies and make it easier to comply with various information privacy and reporting requirements around the globe.
(5) When building cloud-native applications, enterprises should take the PaaS path
The tactical part of the enterprise transition to the cloud entails accommodating existing systems where it makes sense, the strategic component involves building next generation applications around cloud-native services. While a fully custom, DIY approach might sound technically elegant, few organizations have a base of cloud-savvy developers and infrastructure designers of the type of service providers that appear in an AWS commercial. Instead, as I argued in this article, most organizations are far better off using a PaaS stack as an application foundation, particularly an infrastructure-agnostic platform like Cloud Foundry.
PaaS is the smallest of the three as-a-service segments primarily because it is the most disruptive, failing to easily slot into well-known IT product categories like infrastructure and software services. As I discussed:
In contrast, PaaS entails rethinking core software development and IT management processes, often under the rubric of DevOps. The PaaS case requires a level of strategic and cultural buy-in that normally hinders rapid adoption. PaaS is a manifestation of the maturation and ‘mechanization’ of IT services and software development in which infrastructure implementations are rentable commodities and competitive differentiation lies in higher-value digital services. By streamlining the production and maintenance of such services without compromising creative innovation, PaaS is a natural evolution of software methodologies that exploit the advantages of cloud services.
As the article documents, PaaS offers significant benefits via improved developer productivity, with more time to focus on software innovation and less on programming overhead, faster release cycles and better software quality. While each of the major cloud services has PaaS products, there is a significant risk of lock-in, which explains the growing popularity of multi-cloud stacks like Cloud Foundry, Platform9 and others. Expect a growing parade of enterprises to base future cloud-native applications on a PaaS as the costs, risks and learning curve of DIY cloud stacks become apparent.