VMworld Europe - VMware wants to give every user the 'easy button' to press
- Summary:
- A potentially important acquisition in the container world, and a start at building infrastructure tools to make Blockchain easier to work with.
Only a handful of weeks have passed since my colleague Kurt Marko was reporting in from VMware's annual set-piece conference in Las Vegas, but this week in Barcelona its European compatriot event was adding more depth and colour to its push to become the multi-cloud `abstraction layer' for any and all users looking to step into the multi-cloud world.
There was one significant development to emerge there - the acquisition of Heptio – which was finalised the day before the European edition of VMworld opened. In addition, some of the tools announced in Las Vegas in September have now been released out for beta testing, such Dimension and Blockchain services.
The key factor here is that the company is seriously on the hunt for tools that can contribute to building a comprehensive multi-cloud infrastructure `glue' that is aimed at letting users work with extended multi-cloud environments, whether in the public cloud, on-premise, or private cloud operations. The mantra is any device, running any application, on any cloud service, and to deliver that the company is pitching to create a single environment that any application and enterprise might want to use can be written to work with - and to be fair there are already many to which that already applies.
Tim Hearn, VMware's Director of VMware Cloud on AWS had a simple explanation of the goal - and yes, it is a bit glib and something of a marketing soundbite, but it might just sit well with CIOs under pressure from the C-level about the benefits of operating in a multi-cloud environment.
The goal is: every workload on the most appropriate cloud or service for its logical requirements, the priority level of the process and its physical location in the world - and conflicted by the workload and attendant risks of engineering such an environment into a working whole from the ground up.
Or as Hearn put it:
The aim is to give users the opportunity to press the 'easy' button.
Buying, and building, the glue ingredients
Against that background the acquisition of Heptio makes some sense. VMware has already been making inroads into the container world, particularly with Kubernetes in mind. Kubernetes and the movement of applications around a multi-cloud are two sides of the same subject - the former is the way to make the latter happen. In practice this can be easier said than done.
This is where Heptio comes into the picture. Kubernetes is already accepted as the way that containers are deployed and managed, and, in the view of VMware CEO, Pat Gelsinger, the best way to run them is in VMware's reason for existence, virtual machines. He now wants to see VMware accepted by users as "the dialtone for Kubernetes".
The company has already been working with Pivotal to deliver PKS, the Pivotal Kubernetes Service covering customer use cases for on-premises deployment and as a cloud service. This provides the infrastructure tools to allow organisations to run applications in a cloud-agnostic fashion using Kubernetes. Once the acquisition is complete Heptio's Kubernetes management tools will be added into the VMware management infrastructure, focusing on products that make it easier to manage multiple clusters across multiple clouds.
One of the keys here is that Heptio was founded some two years ago by Joe Beda and Craig McLuckie, who were also two of the creators of Kubernetes. As Gelsinger said:
This is like capturing the kings of Kubernetes.
The Heptio acquisition follows a similar pattern to VMware's August move for CloudHealth, a company that provides a multi-cloud management platform that already works across AWS, Microsoft Azure and Google Cloud platforms and gives users a single interface to managing cloud cost, usage, security and performance.
Both fit in with the objective of providing an abstraction layer capable of providing the business process 'glue' between different cloud - and non-cloud - services.
The latest announcements about the company's endeavours in the world of Blockchain also match that criteria, despite the fact that applications are still thin on the ground. But part of that reason, in Gelsinger's view, is the difficulty - or sheer nuisance value - of having to engineer the environment in which it can operate. Such applications are also said to be increasingly energy consumptive. VMware therefore has seen this as an opportunity to develop some appropriate management and infrastructure tools to give users some commonality for Blockchain applications across both applications and cloud services.
This service has just gone into beta testing, with IBM Cloud being one of the first to come on-board. The aim is to provide permissioned Blockchain operations for enterprise consortia with the goal of making transactions intrinsically more secure than public Blockchains. It intends to be the foundation for decentralised trust while at the same time delivering enterprise-grade scalability, reliability, security and manageability.
VMware is partnering with Dell, Deloitte and WWT on the development and support of the new capability, which will focus on the management of day-two operations, supported by the company's availability as a managed SaaS platform with a single pane of glass management interface and enterprise monitoring and auditing tools.
The Project Dimension infrastructure set out by Kurt previously is also now being made available as a beta programme, with the aim of working with companies looking to consume VMware's Cloud Foundation, delivered as a service, right out to the edge, where Pulse covers IoT applications and Workplace One covers people-oriented tasks.
The goal is to deliver the simplicity, agility and velocity of cloud-managed Infrastructure-as-a-Service and physically bring it down to wherever infrastructure is required. Project Dimension will be able to power a host of edge, data center, and hybrid cloud use cases.
Cloud? Got lots of that
AWS is now the centrepiece of VMware’s network of cloud service provider partners, but is in fact only one of some 4,000 listed on the company’s website as verified VMware Cloud providers. Most are numbered amongst the smaller, more regional or niche market providers. But in practice, these will provide many users with the range and depth of choice they are likely to require. But there are many of the bigger names, such as OVH, Claranet, Rackspace and Virtustream available, as well as the other three players in the global quartet of cloud service providers, Microsoft Azure, Google, and Alibaba.
The growing abilities at providing multi-cloud flexibility in consuming containerised applications and services using Kubernetes is expected to play a big part in leveraging this range of options.
VMware expects to have VMware Cloud available at all of the AWS locations by 2020.
IBM Cloud is also one of those bigger names in service provision and was the subject of particular attention at the event. Arvind Krishna, Senior Vice President, of IBM Cloud, told delegates that the company now had some 1,700 clients, with 400 in Europe and, in response to the inevitable question, suggested that the acquisition of Red Hat would accelerate the take-up of hybrid cloud in a multi-cloud world.
He also told them that IBM is now launching a mission-critical cloud service globally. This will be aimed at enterprises looking for zero downtime and automated failovers and will be exploiting the ability to use VMware and Kubernetes together.
Gelsinger also revealed that VMware is now partnering with IBM on developing services using the latter’s Watson AI capabilities to help improve customer service across VMware support portals, using Watson to communicate with them in natural language. Watson is designed to detect product type and version, analyse issues and match those issues with an expert engineer for faster resolution and a better customer support experience.
Gelsinger also announced a new target market for its cloud services in the run up to the arrival of 5G mobile wireless services as a working proposition. This is the Telco market itself. He claimed that this is currently only 10% virtualised, yet he sees 5G creating a host of new applications and services that can best be delivered via the Telco service providers as an integral part of their services, and that that they now really have to keep up:
5G is going to be one of the largest capital build-outs of the next few years, but it needs a more agile infrastructure, as many new applications will be coming along. In particular, network slicing to allow IoT and autonomous vehicles to work properly will be very important.
He said that VMware already has Telco 60 operators signed up.
My take
My opinion does not differ much from that expressed by Kurt Marko. VMware has pulled together an increasingly comprehensive range of multi-cloud infrastructure tools and services and now seems set to go for that market with full malice of forethought. It could stumble over the fact that such changes can often take decades to really emerge and become accepted, so there is the chance it will be left, forlorn, at the altar.
By the same token, there is a market need here that seems clear. Every business leadership is being brow-beaten with the 'go cloud' message, and most now understand the potential benefits of being there, once they have arrived. Getting there is, however, still seen as something of a practical risk for many IT managements, and the suggestion of going multi-cloud will undoubtedly be seen as an exponential expansion of the risks. If VMware can provide an `easy button’ for even some of the necessary engineering required there may well be many-a CIO ready to breath a sigh of relief.
As Kurt points out, there will be competitors coming along very shortly. But if they and VMware work together to build the required bridges between them the whole market will be able to move forward rather than get stifled by unnecessary competition.
As a final thought, if this is as successful as it could be, I would wager a 50p bet on the four big global service providers coming together in some way in order to buy VMware and provide it to all as the unified lingua franca of multi-cloud infrastructure. It would not do them any harm as I suspect most users would love to see `cloud’ disappear as an issue so they can get on with exploiting all the business services that will be available.