The Software Defined Data Center Is an Optimized Cloud Platform
You’re a forward-looking enterprise. You’ve got the newest hardware, great employee perks and you’re moving your IT workload to the cloud — the cloud is really going to help you keep your competitive edge.
Or is it?
IT has struggled with getting the legacy data centers they've been running to manage the new business demands they’re receiving. What technology professionals need is a way to prioritize software over hardware and take advantage of automation options that can release them from doing everything manually.
Software-Defined Infrastructure (SDI) can do those things. And in 2017, when 37% of IT budgets are intended for infrastructure maintenance, you may want to further consider SDI and how you’re going to reach your goals before doing additional spending.
What is SDI?
For a long time, the data center model concentrated on hardware because that was the largest investment and not as easily managed or replaced. But the technology tide, influenced by the cloud, is turning.
With a nod to the B2C application ecosphere model (thanks, iTunes!), the Software-Defined Data Center (SDDC) is an application-first model. SDI is a cloud-based infrastructure that transforms the data center from a hardware-based system to one that focuses on applications and workloads. Application developers are now free to concentrate on building apps in which they define application and operational policies. At the same time, orchestration software automates provisioning and configuration based on the business’s needs.
Figure 1 highlights the features that make up the Intel SDI Vision: self-provisioning, automated orchestration and composable resource pools. Figure 1 also details how the data center of today takes months to provision a new service because of the process that includes evaluating needs and user demands during pre-production, manually configuring devices, and setting up service components and assembling software. On the other hand, software-defined infrastructure takes just minutes to provision new services. The SDI-simplified process includes a self-service catalog and services orchestration, and automated composition of resources and software components to get services running.
The business case for SDI
Intel explains software-defined infrastructure as virtualized, orchestrated and automated resource pools of compute, network and storage. Software-Defined Networking (SDN) allows the network to function more like the virtualized version of compute and storage, resulting in simplified and consolidated administration of networking services. SDN decreases the time needed to create new services and offers real-time optimization of configurations.
Software Defined Storage (SDS) sanctions storage infrastructure management through software, rather than keeping it integrated with hardware. SDS creates a new system of using off-the-shelf, Intel-based servers with automated provisioning to deliver storage optimization. It also helps organizations choose more standard hardware that can save on costs, avoiding expensive and complex proprietary systems.
Network Function Virtualization (NFV) simplifies networks by virtualizing devices so they can be connected. Instead of residing on hardware, like firewalls and routers, the now-centralized, largely standardized, virtualized network is both highly agile and cost effective.
Figure 2 shows different levels within Intel’s software-defined infrastructure. At the top is customer service, where the self-service portal and service level agreements reside. Below that is operations orchestration, automation and intelligence, along with workload health and utilization. The next lower level is the infrastructure, composed of storage, network and servers. At the bottom is the physical facility that includes power, cooling and location.
SDI provides full control of on- or off-premise workload deployments when combined with a cloud-management platform, and completes testing and provisioning in minutes — as compared to what used to take days or months. With SDI, IT organizations can:
- Build a unified services catalog on their infrastructure that enables self-service.
- Secure easy, rapid access to new services and tighter control over workloads.
- Reduce capital expense while increasing operational efficiencies.
- Power all the next-gen services and apps needed to drive new revenue opportunities.
Intel’s SDI model is structured around policy-based software automation processes in three different layers — the orchestration layer, composition layer and hardware pool. Let’s look at the orchestration layer and how it helps the IT department focus more on business priorities without losing visibility or control of how resources are being used.
Orchestration and self-service
Data centers moving to a software-defined architecture can create self-service capabilities that balance IT’s need for control and security with the business’ need for rapid innovation.
Most importantly, SDI allows IT to build a unified services catalog on their infrastructure that enables self-service capabilities for both IT and business users. To stay relevant, organizations need to have this quick and easy access to new services as well as tighter controls over workloads.
SDI, the cloud-management platform also known as an orchestration layer, enables consistent management and policy enforcement across private and public clouds, along with self-service capabilities. IT can build and rely on a centralized, unified catalog of custom and third-party services, many of which can then be combined using policy-driven automation. This enables much faster completion of work streams and extends the benefits of self-service to business consumers. This adds value on several levels, including dashboards that can provide access to the applications, services and even third-party solutions users need.
Orchestration software improves the ability to manage everything from uptime and performance to security and compliance across hybrid cloud environments. It achieves this by enabling resource provisioning and management based on the unique requirements of an application. And, a fully orchestrated SDI also has fewer deployment bottlenecks than public cloud offerings, which can bring costs in line with public cloud for many types of workloads.
Over time, IT can use orchestration technology to build a catalog of standardized application and middleware services with clearly defined costing considerations for use across the business. A self-service catalog will not only help business decision-makers address pressing challenges faster, it will free up time for IT to focus on strategic projects.
The combination of self-service capabilities in a hybrid cloud environment is not the solution for every business challenge; in some cases, traditional IT approaches will still make sense. But increasingly, across all kinds of use cases, self-service provides an unprecedented opportunity for breaking down barriers to innovation and responsiveness. And it not only gives IT more control over managing infrastructure, it can also help strengthen the strategic relationship between IT and line of business decision-makers.