Key Elements of a Modernized IT Infrastructure
Updated solutions help organizations achieve their business goals.
IT modernization is a priority for organizations in every industry. Many executives and IT leaders are finding out that maintaining outdated legacy systems can pose serious risks for their organizations, such as business delays, security issues and subpar user experiences. Conversely, modernized solutions can improve security, reduce maintenance and other costs and free up IT staff for high-value projects.
Given the risks of maintaining outdated systems and the benefits of upgrading the technology environment, its no wonder that 88 percent of businesses expect their IT budgets to grow or stay steady during 2020, according to Spiceworks.
There’s no single roadmap for IT modernization. Every organization’s IT environment is unique, and companies should base their investment decisions on their specific business goals rather than trying to chase trends or replicate other companies’ successful efforts. That said, there are a number of modern storage, networking, compute and cloud technologies that organizations should at least explore as they set out to modernize their IT infrastructure. These solutions tend to provide heightened performance, greater simplicity or improved ease of management for IT administrators.
Modernizing on-premises storage solutions can improve application performance and streamline management tasks.
Software-defined storage: The Storage Networking Industry Association describes software-defined storage as storage virtualization with a service management interface, including automation, standard interfaces, virtualized data path, scalability and transparency. SDS emphasizes interoperability across hardware and software solutions, giving organizations increased flexibility to consume and explore different physical data storage options and improve their ability to generate business insights.
Hyperconverged infrastructure: While hyperconverged infrastructure also incorporates networking and compute, it is the first foray into SDS for many organizations. In addition to the other benefits of SDS, many organizations are attracted to HCI because it is modular and scalable. Many businesses start with just a few nodes in their data centers, then gradually expand their HCI clusters as demands increase.
Flash storage: Data center administrators have long coveted the performance of all-flash arrays. The major difference today is that these solutions have become considerably less costly in recent years, making them viable even for organizations that exclusively considered spinning-disk storage during their most recent infrastructure refreshes. The emergence of Non-Volatile Memory Express (NVMe) is another important development. This high-performance protocol offers lower latencies (even compared with previous flash solutions) and can enable new applications and capabilities for real-time workload processing, both in the data center and at the network edge. Organizations preparing to update their storage infrastructure should look closely at the total cost of ownership for all options, as flash arrays can sometimes provide improved performance at a lower overall cost than spinning-disk solutions.
A robust network is essential for supporting bandwidth-intensive use cases and mobile applications.
Software-defined networking: Software-defined networking solutions separate a network’s data plane from the control plane, making the control plane independent and leading to a more programmable network. Among other benefits, SDN enables organizations to deploy applications and services more quickly by leveraging open application programming interfaces.
Wi-Fi 6: The latest generation of wireless connectivity, Wi-Fi 6 (also known as 802.11ax) offers a theoretical maximum speed of 9.6 gigabits per second, a dramatic increase from the 3.5Gbps of the previous Wi-Fi standard. Because this is far more speed than any one user or machine is likely to require, the benefits of upgrading to Wi-Fi 6 come primarily from improving overall wireless network performance.
Intent-based networking: An intent-based networking solution allows network administrators to define a desired state for the network, and then uses automation to implement these desired (or “intended”) policies. With this approach, data center administrators can ensure sufficient performance for specific traffic, simplify management and maintenance, and improve security.
Network management: By adopting network management software, organizations can reduce complexity and improve productivity, resulting in a lighter management burden for IT staffers. This can lead to lower networking costs, accelerated time to market for new services and an enhanced end-user experience.
Aging compute infrastructure is a major source of inefficiency in enterprise data centers. According to IDC (PDF), organizations that replace their servers see a substantial reduction in costs associated with maintenance, power and facilities over a three-year period and significantly reduce their unplanned downtime. (According to some estimates, the cost of data center downtime alone can approach $10,000 per minute.) In organizations that leave aging servers in place, meanwhile, IDC reports that IT staffs spend more than half their time conducting routine maintenance tasks. Server upgrades are also associated with a reduction in total physical infrastructure footprint, an increase in virtual machines per physical server, and faster development and deployment of new applications. Yet many organizations try to squeeze five, six or even seven years of life out of their compute infrastructures.
IT administrators have a number of options when upgrading their compute environments. Intel has long been a driver of innovation as it has advanced processing according to Moore’s law. The vendor’s latest move forward is a new memory module, Optane. Based on Data Center Persistent Memory Module, Optane can load frequently used database tables and rows into memory for significant performance enhancements. Alternatively, AMD’s EPYC chipset enables many more virtual machines than a standard chipset.
Many organizations have moved away from the cloud-first mindset that once prevailed, instead evaluating the public cloud on a case-by-case basis. Still, it is important for organizations to do their due diligence by exploring whether portions of their environments could run more efficiently in the public cloud. Not only does public cloud infrastructure support scalability and agility, but public cloud resources also require far less maintenance and support than on-premises resources. In a sense, infrastructure in the public cloud is always “modernized.” But to get the most out of the public cloud, IT administrators must take care to appropriately design and optimize their environments.
Cloud management platforms are essential for helping IT professionals to gain needed visibility into and control over their public cloud environments. Governance policies, often established by a multidepartment cloud center of excellence, are also important for preventing the sort of sprawl that can quickly lead to cost overruns.
Hybrid clouds that incorporate both private and public resources have become the norm in enterprise data centers, and organizations are increasingly looking to multicloud environments that tie together on-premises resources with more than one public cloud. This approach not only allows organizations to take advantage of different vendors’ strengths and unique features, but may also put them in a stronger bargaining position when negotiating new contracts with cloud vendors.
To learn more about the opportunities presented by an updated IT environment, read the CDW white paper “Meeting Demands for a Modernized IT Infrastructure.”