Home > Blogs > Rethink IT > Tag Archives: DevOps

Tag Archives: DevOps

Shifting from Infrastructure to Innovation

In my conversations with CIOs and other IT executives, I often hear how their teams are focused on maintaining a solid, reliable infrastructure. Their priorities are continuity of service, meeting SLAs, and minimizing disruptions and downtime. That’s an important, admirable goal, but as every IT exec now knows it’s not the whole picture.

If your teams spend too much energy on maintenance to ensure things don’t go wrong, they’re probably going to miss the opportunity for moving forward – and the threat of being left behind.  Consumerization of IT and the cloud have changed everything. As one customer exec pointed out to me recently, “Public cloud options can be the pink slip for IT infrastructure and operations teams.” Let’s face it, the monopoly is over.  Public cloud services, both consumer and business, have set a new standard for IT service delivery – ease of access, speed, reliability, etc. – as well as expectations on price, and IT teams are expected to match or better that standard if they want to stay in the game.

With so much available today on demand in the cloud there’s greater pressure than ever on IT to somehow reduce expense and shift Keeping the Lights On to new, innovative projects that drive business productivity and profit growth. You need to empower your teams to think and act differently, enabling them to be a world class IT organization.

Your teams can no longer focus on the infrastructure; they have to focus on taking advantage of the infrastructure to deliver new business value through innovation. In a world of options – private cloud, public cloud, hybrid cloud, virtualized and physical infrastructures – the focus needs to be on making the right choice that’s right for your business.

The question is no longer “How do I make my infrastructure the best it can be?” but: “What’s the best infrastructure for what we want to do?”  IT has to decide the most logical place to provision and operate infrastructure and applications, based on criteria such as cost, risk, compliance, security, etc. That’s where the innovation comes in – what works best where? What capabilities can I start to deliver as services? What cloud services can I take advantage of to help drive what the business is trying to achieve?

That’s the shift we’re seeing in IT. Instead of providing a super reliable infrastructure to support your applications, you’ll be sourcing and providing services. As I mentioned in my previous post, IT will become a broker for services that the business needs, with a fact-based approach to identifying the best source of those services, internally or externally. Being a service broker will help your teams shift toward innovation, while matching or bettering the standard set by public cloud services.

Some of those services – the ones supporting your mission-critical activities – will stay on-premise for reasons of security and compliance. Some of them – the utility part – you’ll offload to a cloud infrastructure provider through IaaS. The rest of them – the part in the middle – you’ll offload to a SaaS or PaaS vendor (someday these may come back in house or they may stay in the cloud or even move back and forth depending on cost and changing business demand).

Being an innovative IT organization is about trying new things. About being daring. About making decisions faster, killing projects sooner, investing more in projects that warrant it. And about how the cloud – private, public, hybrid – can help you do that.

This is going to take a mind-shift on the part of your teams and a critical look at your processes. You’re going to have to be more customer-centric and deliver cost transparency to your stakeholders. You’re going to have to standardize the services you offer (think 80/20 rule) and enable self-service access to them. And you’re going to have to put the right governance processes in place – who gets access to what and where does your data live.

In my next post I’ll walk through how you can tackle these challenges.

Care to comment on this blog post? Share your thoughts your thoughts with us in comment section.  

Delivering Business Innovation with Application Management A presentation at the O’Reilly Velocity Conference

A few weeks ago at the O’Reilly Velocity conference, Komal Mangtani, who heads up engineering for our Application Management business, and I co-presented VMware’s point-of-view on how applications need to be managed in the cloud era and how you can leverage the cloud to drive business agility and operational efficiency for your IT organization. Our presentation can be viewed here.

Velocity attracts many cutting edge, “New Age” companies as well as big, established players. Attendees are mainly system administrators but there’s also a large contingent of developers and operation teams.

Komal and I kicked off our presentation by walking through a number of examples of how business and IT innovation have been accelerating over the past few decades. We made the following case to the audience:  not only is the speed of innovation accelerating, but it’s fundamentally changing the way applications need to be built, deployed and managed.

As we analyzed these examples, it was clear to everyone that software is the key element driving innovation today. Marc Andreessen made this very point last year when he wrote in the Wall Street Journal that every company is a software company. It doesn’t matter what industry you’re in – Manufacturing, Telecom, Financial Services, Pharmaceuticals – what you’re doing is powered by software.

Just a few short years ago, IT was a business enabler; today, IT is the business. While most business and IT executives recognize this truism, many of them are unsatisfied with their innovation performance. The key reason is that the innovation delivery chain suffers from a split personality disorder:  developers want to innovate with abandon, while operations teams want a very stable environment with tight control. For ages developers have been seeking ways to improve their software development lifecycle, moving from waterfall to agile and scrum methodologies in order to get their latest and greatest code rolled into production as quickly as possible. Operations teams, however, have put in strict rules to slow down that process in order to maintain control and avoid outages. With these two forces pulling in opposite directions, innovation can’t thrive and IT developers and operations teams – and, more importantly, the business – suffer.

One of the latest trends that is closely tied to rapid innovation and execution is the adoption of Platform as a Service (PaaS). Admittedly, PaaS is an over-hyped concept, but in general PaaS does indeed simplify the development process and enable ongoing innovation. However, it’s still a maturing approach and many solutions in the market lack significant enterprise features while threatening to create vendor lock-in.  In fact, one of the core distinctions VMware’s “Cloud Foundry” has delivered is openness with respect to where the application can get deployed, in recognition to the lock-in challenge.

In our talk, Komal and I differentiated between various types of PaaS. Although we use the word PaaS very commonly now, the industry has recognized that there is more than one PaaS. There is aPaaS  (Application PaaS)– where solutions like Google AppEngine, force.com, Intuit serve as examples and there is iPaas (Integration PaaS) –  examples include IBM CastIron and Informatica Cloud Services.

Some PaaS solutions are built exclusively to users for a specific SaaS application like NetSuite BOS while others are independent like LongJump or Relational Networks. And so, we highlighted that when talking / exploring PaaS, make sure you understand the full context and extensibility of each of these offerings.

The reality, however, is that it’s still early days for PaaS adoption and most business applications do not, and will not run on PaaS in the near future. The industry generally agrees that PaaS will be the application platform of the cloud era and a very different developer experience, one where they can focus on core development and not environmental details. Hence, the driver for PaaS, the ability to deliver applications faster to market, needs to be balanced and tackled per project as it is still not a one size fits all.

The $64,000 Question (for those of you who remember that iconic game show) is, what is? The answer lies in understanding the nature of the challenge.

In order to enable innovation today and to make an impact on your business, you need to be able to set up new environments quickly. You need to be able to service millions of users around the clock. You need to be able to develop new capabilities in an agile manner and apply these changes on an ongoing basis. The evolution of IaaS has made getting infrastructure up and running in minutes a reality. At the same time, agile development has had a huge impact on making newly developed software available much more quickly. The challenge today is to get that newly developed software up and running on the instantly available infrastructure in a controlled manner while maintaining agreed service levels.

In my next post, I’ll walk you through how the VMware Application Management Suite addresses this challenge, and how you can deliver innovation for your business by bridging your development and operations teams to achieve agile operations.