Hi. Welcome to Ingram Micro.

Please choose your role, so we can direct you to what you’re looking for.

If you’d like to learn more about Ingram Micro global initiatives and operations, visit ingrammicro.com.

How Data Center Requirements Have Evolved over 5 Years

December 21, 2017

How Data Center Requirements Have Evolved over 5 Years

Data center design continues to evolve with the adoption of new technology, and the technology that has had the biggest impact on data center requirements in the past five years has to be cloud computing. Rather than reinforcing firewalls to keep out intruders, IT managers are more focused on creating virtualized infrastructures that integrate cloud and remote resources into a single, secure system. The evolution of the data center requires the IT department to look beyond the network that is within its immediate control to securely integrate external systems and applications.

Worldwide, cloud adoption continues to grow at a robust rate. Gartner, Inc. predicts that cloud use will climb 17.2 percent by the end of 2016 to $208.6 billion, up from $178 billion in 2015. Software as a service (SaaS), which remains one of the largest segments, is expected to grow by 21.7 percent to $38.9 billion, but the highest growth will come from infrastructure as a service (IaaS), which is projected to grow by 42.8 percent in 2016.

Clearly, cloud computing is driving new data center requirements to accommodate solutions such as IaaS. It’s also driving adoption of related solutions, such as hybrid clouds, software-defined networking and virtualization, which necessitates rethinking data center requirements.

The Growing Impact of the Cloud

One of the reasons why cloud computing has become so attractive to CIOs is its reduced cost and increased flexibility. Rather than requiring investments in hardware and enterprise software, cloud computing offers a more cost-effective approach with no additional hardware requirements and no new enterprise software, as well as scalable computing and data storage space. Plus, it can be deployed in a fraction of the time needed to expand the on-site enterprise.

Public cloud systems have proven ideal for testing new services and new applications (e.g., SaaS), and as the cloud continues to provide value, organizations are adopting more private and hybrid cloud services for their computing needs. Private cloud systems provide the same efficiency as public cloud systems (e.g., fast deployment and scalability) but with more centralized control and greater reliability. Some argue that private clouds are more costly, but when you take into account the additional cost in hardware and software to support the same kind of computing in-house, cloud computing can mean real savings. Hybrid clouds offer the best of both, offering the flexibility and scalability of public cloud systems but also the control over application-specific private cloud resources. The hybrid cloud is increasingly providing a foundation for IaaS as more companies rely on cloud computing for business-critical applications.

Cloud computing adoption affects data center requirements and demands changes in the physical architecture. For example, new kinds of servers are needed with more efficient rack systems and data storage. Cloud computing also requires more bandwidth and high availability, which means more load balancing and traffic management for network data. Hybrid cloud systems have to balance on-premises computing resources with colocation options, as well as managed services. And cloud applications need to be assessed for their suitability for integration and interoperability.

Software-Defined Network Simplifies Routing

Software-defined networking (SDN) is becoming increasingly popular to support cloud computing. With SDN, the hardware and software dependencies are separated by abstracting the lower-level control functions. Controls for switching and other functions are placed on a higher plane that can be controlled in software, which makes it easier to manage data processing and network load balancing.

More important for data center design, SDN provides more efficient data routing. SDN allows IT administrators to centralize control without having to manually configure hardware or systems, including cloud services. This makes it easier to implement rules to prioritize data traffic for better performance. It also enables greater security by supporting different security layers, creating secure data pipes and encrypting data.

Virtualizing Network Resources

Virtualization also has had an impact on data center requirements. Server virtualization, which divides a physical server into multiple virtual environments to consolidate server resources and data storage and to improve efficiency, is expanding to support network virtualization, which provides the same type of abstraction but creates smaller subsets of networks within a single infrastructure. In short, network virtualization allows multiple networks to exist on the same physical network.

Network virtualization promises to improve data center efficiency while reducing costs. The technology also can run on high-performance x86 machines and provides a software-defined data tunnel rather than a physical connection to connect two domains. Savings come from the fact that the IT team doesn’t have to implement new wiring or hardware; instead, it can virtualize services and configure them in software.

Virtualization extends to cloud resources, as well as data center resources. Anything in the infrastructure can be delivered as a virtual service. This also simplifies data management, and better data center management tools are emerging for virtualized resources. Once you separate the control information from the actual data, traffic management and tuning using software are much easier.

As you can see, cloud computing is changing data center requirements in dramatic ways. In a nutshell, the data center is becoming less a center for actual data processing and storage and more of a traffic control center to integrate virtualized resources on site and in the cloud. Performance is being measured in how fast data can be abstracted and processed rather than just the speed of actual data processing. The more resources migrate to the cloud, the more emphasis will be placed on bandwidth, load balancing and streamlining data traffic handling to optimize data center efficiency.