on 10-10-201603:06 PM - last edited on 10-20-201602:31 PM by jason_cmgr
10 years ago, when I got my first smartphone, a blackberry, from my employer, I was super excited. Things have changed dramatically since then. The pace of technology innovation across Cloud, Mobile, Social, and Big Data have disrupted industries and transformed user experiences. Today, I choose the device I bring to my workplace. I have a choice of applications to collaborate with my team and with my customers. The devices and applications have become platforms for new types of business and user innovation.
As organizations go through this digital transformation, app developers responsible for these digital experiences, are choosing where and how to develop their apps, and having a big influence on next gen IT and data center architectures. For time-to-market and business agility, application are moving to cloud for economies of scale and ease of use. According to IDC’s November 2015 CloudView Survey,58% of all organizations are embracing the cloud, and 43% of enterprises expect majority of their IT capability to be delivered via public cloud services. However, cloud adoption means applications are designed to scale across multiple data centers and cloud.
The leaders of hyperscale data centers – Facebook, Google, AWS, emerging players like Netflix, eBay, Walmart – have demonstrated that irrespective of where you build and host your applications, your data center design strategy and IT culture has a huge implication on business agility, growth and customer experience. Virtualization of compute, storage and the network improved utilization and scale and catalyzed the move towards software driven data center design. The earliest adopters of this trend, webscale IT, also used software automation, orchestration and management to reduce human intervention and increase operational efficiency. This would not have been possible without a change in their IT culture. Initiatives, like DevOps, Microservices architectures, workflow and cross-domain automation across IT silos, enabled simplifying otherwise complex inter-connected data center designs.
What does this mean for you as the IT leader, for your data center architecture and IT culture?
The many approaches to data center design strategy can be confusing. But, essentially there are 2 approaches – a vertically integrated complete stack solution or build your own stack. Each option comes with tradeoffs.
Complete stack solutions might be easy to deploy, your IT team may already have all the skills sets to manage and operate it. But in the long run, this data center design approach is not scalable. You may not be able to choose, redefine your application architectures, or move to a multi-cloud environment as cloud adoption continues. If your goal is to move your IT organization, to an IT-as-a-delivery services organization, then an integrated stack or vertical solution may not be the answer. In addition to abstracting the complexity, the lack of visibility into your data center stack, can have serious implications on trouble-shooting, flexibility to scale and impede progress in the long run. You will still need to invest in additional tools, processes to integrate with rest of your management, operations and orchestration systems. So overall CapEx and OpEx will actually increase.
A custom stack or build-your-own-stack approach, in the short term will require investment of engineering effort for data center design and integration. In the longer term, it gives flexibility to and huge cost advantages. A custom-stack approach also enables you to break down your IT silos as teams across storage, compute, network, application developers and IT operations collaborate and contribute to the data center design. There is a parallel here with bring-your-own-device and bring-your-own-app.
In a custom stack approach, you can modularize the elements of your data center stack into building blocks. The most important ones are Orchestration & Management, Control, Software and Hardware. Embracing this approach, means embracing purpose-built composable solutions that can be integrated into a complete data center stack for your specific functions, applications and to address your specific business need. This approach enables maximum flexibility. As you adopt more open source and open standards based software or hardware solutions, you can further improve cost efficiencies. In addition, composable solutions built on open standards, and integrated with an open ecosystem, are readily interchangeable without a fork-lift upgrade or large-scale data center design overhauls.
The network plays a huge role in this ever evolving distributed cloud computing architecture, connecting data, applications and the end users. And in hyperscale data center design, the network is a crucial element of their architecture as well as software based workflow automation, orchestration and management. In a recent Forrester paper on networking vendor selection, Forrester recommends enterprise networking teams to choose their networking vendor and networking solutions based on features they actually need. Too often, IT teams make decisions based on comfort level, brand awareness and features which they may never use.
At Brocade, we aligned our strategy to help our customers build their next generation data center design and integrate multi-cloud environments. As leaders in data center storage networking, our expanded portfolio includes purpose built, composable, high performance data center networking hardware, software, and automation solutions. Built on open standards, integrated with open ecosystem of open source as well as leading solution vendors in the data center space, we enable our customers to compose a data center design best suited for their digital transformation.