Extending Compute Virtualization To All Applications
Whilst many consider the compute virtualization market to be relatively mature, the story is far from complete. As the technology has become more pervasive and mature, there is an enormous drive to use it to provide an efficient, flexible, unified infrastructure pool for all applications, including back office, mission critical, big data and high performance computing applications. Our vSphere releases in 2013 reflect this focus, with improvements in ESX scaling, support for more processors, and extensions for big data. Our sense is that in 2014 we will see a significant uptick in the use of virtual infrastructure to host big data and grid applications – as enterprises seek to maximize the work they can do, and make application deployment more flexible and more reliable. We should also expect a continued focus on Big Data and HPC as the IT enablers of differentiation in business. In the Cloud-Mobile era, differentiation is driven by intimacy, i.e. by analyzing big data, and through simulation, i.e. HPC.
Network Virtualization
The brouhaha around Software-Defined Networking that started in 2012, and extended into 2013, will not abate in 2014. We expect to see more of a focus on network virtualization deployment next year. VMware acquired Nicira in 2012, and in 2013 we announced VMware NSX, the platform for network virtualization. NSX delivers a layer of software that separates, or abstracts, network distributed applications from the underlying physical network. This delivers many of the same benefits that ESX delivers to applications within the compute domain – i.e. improved efficiency, much greater agility and greater reliability, as operations become more automated.
Storage Virtualization
The separation of applications from their storage is ongoing. In 2013 we acquired Virsto, a small company whose software sits between the applications and storage to help manage performance, and accelerate the public beta of Virtual SAN (vSAN). As we bring these and other storage efforts together with our software-defined storage strategy, we will deliver the layer of software that allows us to make data easily available within VMware infrastructure, and to make it easily consumable by applications, irrespective of the underlying physical storage infrastructure. This trend toward abstracting storage is gathering pace across the industry, and we expect to see significant progress in 2014.
Automatically Managing The SDDC
In spite of the amazing progress being made in separating applications from the underlying compute, network and storage infrastructure, this is only half of the SDDC story. If we think in terms of a life cycle of an application, we need to automatically map it onto the underlying infrastructure to deliver the right service level attributes (performance, scale, throughput, availability, security) and cost. We need to monitor it. And we need to close the loop and act on monitoring information to remap, or re-provision the application automatically, if it is not meeting its service level objectives. And finally, we need to provide feedback to the owners of the applications and the infrastructure, around cost, capacity planning and so forth. This is the purpose of VMware’s Cloud Automation, Cloud Operations and IT Business Management products, all of which have evolved significantly. The broad trend seems to be towards managing heterogeneous SDDCs (and thus Clouds) so that consumers of *aaS are able to manage a broad portfolio of SaaS, PaaS, IaaS providers, together with their private Clouds and partner service provider assets.
One of the fascinating aspects of management is that with IaaS there is a level of opaqueness. Applications should not care about the infrastructure beneath them, and the IaaS owners will generally not understand the semantics of the applications as they do not control them – IaaS is self-service! So we have to use machine learning and big data to infer structure, and good and bad behavior. Machine learning lies at the heart of vCenter Operations Manager, and Log Insight uses big data techniques to evaluate log files. One of the things I would expect to see across the industry in general in 2014 is more use of these techniques, and tying these to provisioning engines, to enable more automated, policy driven closed feedback loops, for application service level management.
Cloud Computing
The hybrid cloud will be big in 2014! Our announcement this year of the delivery of vCloud Hybrid Service, a VMware-operated Cloud, addresses the strong desire our customers have for a public cloud service that looks like their own private clouds, yet has all of the traditional cloud benefits such as greater agility and flexibility. This is just one indicator of a growing recognition that the “New IT” will be about managing a portfolio of services and assets that are on-premise, off-premise, owned or rented. And that consequently, IT will trend towards being essentially hybrid.
End User Computing
Just as the delivery of applications has been transformed by the cloud business model and the software-defined data center technologies that underpin them, the client is undergoing enormous change. The rise of mobile computing is unstoppable. These devices not only enable simple access to applications anytime and anyplace, they also allow those applications to be massively personalized in real time, based on the capabilities of the client device itself – location, video capture, and audio capture. The end of 2012 saw the sales of smart devices (smart phones and tablets) overtake the sales of PCs and laptops. And in 2013 we saw the total installed base of smart devices overtake PCs and laptops. When you add the fact that the majority of all new applications are now written to run natively on either Android or iOS, it becomes pretty obvious that how we consume applications and data is in the process of radical change.
Final Thoughts
If there is one conclusion to be drawn from all of the above, it is that we are slap bang in the middle of incredible change. Whilst change is perpetual in IT, a set of technologies – virtualization and automation in the form the software-defined data center – have enabled a new business model: the cloud business model. The notion of frictionless IT, that cloud makes real, is transforming everything. And cloud has reset everyone’s expectation of what IT is. When combined with the mobile client revolution, I would argue we are working in perhaps the most exciting of times. And 2014 promises to be another year of disruption and opportunity.
