Anterovium - Fotolia
Predictions of rapid-fire growth in the adoption of cloud computing are coming to fruition, with an astonishing number of large- and smaller-scale projects on the docket. Enterprises are flocking to the medium -- not just as a way to cut expenses and help them scale better, but as a means to become more agile and competitive. The expectation is demand -- and cloud traffic -- will only continue to soar. Cisco, for example, said more than four-fifths of all workloads will be processed by cloud data centers by 2019, with global IP cloud data accounting for 83% of all IP traffic.
At this fast and furious pace, there is a real concern that organizations racing toward the adoption of cloud computing aren't taking a methodical approach to updating their network infrastructures. Given the tremendous growth in hybrid clouds -- where workloads are moved between and among multiple clouds, including on-premises facilities -- not having a network optimized to support cloud traffic can be an unfortunate oversight that could impede performance and introduce instability.
Adding bandwidth is only one consideration; IT also needs to consider how design elements and execution strategies may help or hinder efforts to get the maximum benefit of working with the cloud. Among those considerations:
- Does network gear optimize traffic based on the type of workload?
- Is network infrastructure robust, and does the organization have tools in place to ensure network stability and security?
- Does the network provide a consistent and high-performing experience for the end user?
- Has the organization committed to emerging technologies, such as software-defined networking, to enhance its investment in the cloud?
Better by design
One of the keys to having a network that facilitates optimal cloud performance is manageability. This can be a challenge, as organizations look to connect and migrate on-premises application workloads to resources in off-site clouds. From an infrastructure perspective, enterprises are looking to streamline their networks to get a better handle on the environment.
Employing techniques such as implementing a flatter network design -- which removes the aggregation layer and applies a one- or two-tiered architecture -- can simplify both planning and administration. Removing a switching tier allows direct connectivity between servers and nodes, thus reducing latency and shaving operational support expenses.
Many businesses also opt to segment their networks along business lines. By taking this approach, each business line gets the computing and application resources it needs.
Service management and security
IT execs must also have the right service management tools, so they can track usage and administer resources across their environments. Providers such as ServiceNow can meter usage across multiple locations and the cloud. These tools employ automation to support dynamic resource allocation as capacity needs fluctuate. Service management options also ensure the departments using the network are charged accurately and efficiently.
Security is critical -- not just to ensure cloud resources are protected, but that they are also available. The virtual and highly distributed nature of cloud computing obscures visibility and makes identifying and isolating threats challenging.
A new crop of cloud security tools has emerged that address not just the visibility challenge, but also the need for faster detection and mitigation in what can be very complex, sprawling environments. Software like the Incident Response Platform from Resilient Systems -- now owned by IBM -- works with firewalls, as well as other detection and prevention appliances and software, to detect potential issues and initiate a response.
Security products are one part of the equation. Policies and practices are just as important. Organizations must ensure they have crafted effective policies, and then communicate those policies to staff and external partners who may have access to network resources.
All of the security elements, along with service management, are critical to maintaining a stable and highly available network infrastructure that provides the conduit for the adoption of cloud computing. It is crucial that enterprises make the investment upfront in design to support a network that is optimized to carry cloud traffic. As too many organizations are learning, simply adding a bigger switch isn't enough. Instead, IT has to look at network design holistically and consider future innovations before it invests.
Unearth the layers of cloud ⇉
This article is part of a series that breaks down the different technologies that underpin cloud-based infrastructure. Navigate here to see the other articles.
Cloud security tools stop criminals in their tracks
Network modernization in an optical era
Can VPNs secure cloud traffic?