 | Author Bio: Frank O’Connor, Enterprise Technologist, Office of the CTO, has a global outbound focus on the operational impact of Converged Infrastructure with an emphasis on solutions. Frank has a more than 25-year track record, highlighting consistent success in applying strategic vision to merge business with technology. He possesses expertise in business operations, technology integration, network engineering, business relations and overall organizational leadership. Prior to joining Dell Frank was part of the Office of the CTO at VCE, Frank also founded a Homeland Security development company, ESP Sensors, where he built Radiation Detection Systems. ESP provided systems to military, healthcare and public facilities throughout the country. Earlier in his career, Frank served as CIO for Jenny Craig International, J&R Music World and Wickes Furniture. |
What is a Cloud system anyway? And more importantly, why should I care?
First, let’s put some color to this. From a hardware perspective a “Cloud” system is nothing more than a collection of servers, storage and networking, sounds pretty familiar doesn’t it? There are good ways to do this, and not so good ways; I will come back to this in a minute.
From a software perspective things get a little more interesting. Some would say that just virtualizing your compute and storage resources using a Hypervisor and some Software defined Storage controller gives you that layer. Others speak about adding in the virtual networking layer as well; and there is control of the application platform, provisioning servers, storage, backup and load management, namely a catalogue and some self-service portal. Lastly, there is the question of where all of this is, public, private, or hybrid.
So as you can see, the software side is very important, but let’s not forget that like your car, reliability of the components has gotten so good, and the interoperability of the discreet systems so tight, that we start to forget about all of the underlying technology that enable the 40+ MPG in a high performance package.
Many business leaders are asking the basic questions posed in the title of this report, and frankly they are correct to ask. The challenge is the broad array of answers that are proposed, and the level of authority with which so many answer with would make you think there is only one answer when nothing is further from the truth.
As IT has become a greater focus in business transformation, the speed at which new applications and new touch points to customers must be delivered is increasing
The business requirements that demand change have driven the time it takes to conceive of, design, test, and launch an application from what was months, sometimes years, to now weeks, months and sometimes days. The demand placed on backend infrastructure by the Internet of Things (IoT) and by Mobile apps is causing a disruption in the market not unlike Uber to the Taxi Industry, or Tesla in the automotive space.
Consider two examples of how that is already taking effect. At the University of Southern California, Uber is now the dispatch engine for the Campus Cruiser afterhours shuttle. Instead of trying to create their own solution USC co-opted with Uber and Uber created a private slice within their system to accommodate this need. The speed at which this happened was remarkable, and the benefit to the USC community was immediate and easy to manage. In NYC, there are Taxi operators turning off their marquise on top of the cab and signing into Uber; boy, how far we have come. The demand on IT to allow this kind of pivot within a business from an execution perspective cannot be overstated, companies will live and die based on the success and failure of these initiatives.
The engine must also adhere to and embrace standards because, without them, we will not be able to fit into the tightly integrated delivery system that is the Cloud
So having an engine to power all of this is incredibly important. The engine must be flexible, scalable and reliable. Consider that almost all of us are already using the Cloud in some way. At home, with online banking, we may setup and manage access from our desktop computers, but we likely do more transfers and deposits using our phones. Acceptance is immediate and when done from just outside of the bank, one could see the result on the ATM, or ask a teller to confirm. That is the cloud delivering game changing service.
At work when moving from internal applications, HR, to travel (Concur), to Sales Management (SFDC), it is done without regard for where the app is hosted. In fact, users likely traversed at least four different networks without having to sign in a second time, all secured by VPN connection at a remote office location. That is the cloud delivering truly transformational services to business and enabling work, productively, from a remote location.
The speed at which change is embraced in these examples is evident by the number of times your local apps are updated on your phone. In most cases, rarely a month goes by without an update being pushed to your device, enabling a new feature or fixing a bug. Changes are small and incremental, allowing the developers to be very nimble and address needs much faster than just a few years back.
The need for change drives the need for nimble hardware as well
A new app is launched and demand is much greater than anticipated, do your systems recognize the demand? Do you make operators aware of the change and wait? Well not so much anymore. The new systems, the Cloud, now get metrics about demand on the application, can understand the need for more compute or more storage and then make a request to the hardware layer to allocate and deploy these additional systems components, adding the needed capacity to the application to maintain the SLA committed to by the business. Sounds a little like magic, but consider that without a clear and thorough understanding about the nature and capacity of the additional hardware that is available, the add may or may not have the desired effect on performance, and now we are in trouble.
This is the root of the issue. We have all of these tools, Cisco has their InterCloud offering, VMware has VCloud Air, Amazon has AWS, Microsoft has Azure. Some of these are public only, some try to be hybrid, but all require some level of high performance hardware that is able to scale up, and scale out quickly and reliably.
Having a complete set of system offerings that address these needs, while embracing open standards, is a key imperative to the long term success of many of these initiatives
That is where the idea of convergence starts to make sense. The idea that the architecture is able to accommodate additional resources, be it storage, networking, or compute as required to meet demand, is a game changer. The execution of this convergence means that the new hardware can be discovered by a management agent, understood in terms of capacity and capability, and then incorporated into the hardware pool so that applications can consume this hardware as needed, in a reliable and predicable fashion.
Converged infrastructure is the system the enables the nimble enterprise
At Dell, we have built many of the parts to allow customers to realize this vision. We do not do it all, we embrace the standards that allow this to work using our systems together with other leaders in the space. We work well with VMWare, Microsoft, and Red Hat to name a few on the private side. And based on their tool kits, support the movement of applications from public to private environments that truly represent the Hybrid offering so many are trying to reach.
A few companies have a mature vision of what their hybrid cloud strategies are, but no other company can offer the standards based array of products that can be orchestrated into a scalable fabric like Dell can today.
And with our most recent platforms, FX2 and the SC4000, our ability to provide scale up and scale out architectures is second to none. The message is clear, we have built the products. Now we need to help drive the strategy.
So why should you care? You should care because the business demands on technology today will simply crush any IT organization that does not embrace automation within the context of a massively scalable architecture at the enterprise level. By making the correct choices today, by building a foundation based on flexible standardized converged architecture a company can ensure that choices made today will support the business for years to come.