February 17, 2017
Transitioning from Traditional to Software-Defined Data Centers
As software-defined networking and storage move toward full-scale operation, what should IT teams do to achieve SDDCs?
I’ve talked to thousands of IT leaders about the software-defined data center (SDDC) and what it can deliver: cost savings, efficiency gains, consolidation, management improvements and automation. To get to the point where they can reap these rewards, though, IT leaders need a forward-looking mindset, focused on infrastructure and business alignment.
The first glimpse into SDDC’s promise came more than a decade ago, with server virtualization. Next up was software-defined storage, followed more recently by network virtualization, which got a big push forward with the release of VMware’s NSX and Cisco Systems’ ACI.
But the way many IT experts talk about SDDC misses the mark. The focus of an SDDC kick-off conversation should be more about why, not what.
Ultimately, the goal of SDDC is to become service-focused. If CIOs can separate from the hardware and start thinking of their organizations as providers of IT services, they’ll become more focused on the consumers of those services and their specific business needs. And that’s what business leaders want.
Preparing for SDDCs
For presentations to customers and other audiences, I’ve been able to use the same slide deck for eight years — with tweaks here and there to reflect market shifts — because the upfront work needed to prepare for SDDC hasn’t changed.
These aren’t the kinds of projects that are completed in a week or two. Transforming a traditional enterprise data center, or several, into an SDDC — enabling policy-based management of application workloads running in private, hybrid or public cloud environments — takes about three years, on average. CDW works with customers who reside at every point along the IT evolutionary curve, from organizations still running 10 or 20 data centers to those in the optimization phase, testing capacity planning and other advanced capabilities. On the way to the SDDC, we help these organizations navigate several key initiatives.
The two primary areas we focus on with customers are database consolidation and data center consolidation (for both on-premises and co-located data centers). We’ve developed a process for analyzing the workload and computing model across many data centers and, factoring in year-over-year growth, defining the composition of the target data center.
Moving Toward Total Virtualization
Organizations need to virtualize 100 percent of their infrastructures. This goal is difficult for organizations in some industries — healthcare, for one — as some vendors support their applications only on physical machines. These providers are playing a dangerous game of chicken, though, as it’s only a matter of time before some enterprising developer writes a cloud-based version and knocks them out of the market.
This process can be simplified by investing in a hyperconverged infrastructure (HCI) solution. HCI bundles computing, storage and networking — with each provided by a single vendor — in one box.
A line item in this effort is to automate backup through a Disaster Recovery as a Service (DRaaS) provider, such as Actifio. This company and other DRaaS providers manage backup and disaster recovery, as well as another challenge IT teams are facing — controlling data growth in the cloud.
Virtualization and standardization through HCI and application choices are prerequisites to automation. IT leaders should be investing in automation, but it can be complicated, requiring developers that are infrastructure-aware. Still, this is where IT environments are headed. By 2020, I believe, all administrative tasks will be automated.
Self-service, through portals that allow authorized developers, administrators and other users to choose application or infrastructure services (both private and public) will be the single biggest automation win that enterprises see from SDDC efforts. Enabling developers to provision their own tailor-made virtual machines, versus submitting a request to the IT department (which might take six weeks to fulfill), is a win on several fronts.
Focusing on Service
To become service-focused, IT teams should adopt at least some parts of the best practices framework established by the IT Infrastructure Library. The four key areas to focus on are incident management, problem management, change management and the configuration management database.
Managing Your IT Portfolio
Organizations should implement an IT-specific portfolio management system. When moving toward an SDDC, the budget has to cover not only technology investments, but also staffing, so stakeholders need a high-level way to review workforce variables.
An IT-specific portfolio management platform provides IT leaders with visibility into all information related to work within the data center, including staff activities, resource utilization, specific business and IT projects, nonproject work, and unplanned work. The platform should bundle reporting and analysis tools for analyzing different variables related to specific services, delivery cost and the percentage of staff time dedicated to unplanned work — a notorious project killer.
Learn more about how CDW’s solutions and services can help you achieve an SDDC.