network foundations (humphery/


Building a cloud strategy on a solid foundation

Abruptly moving from legacy systems to the cloud is akin to building a brand-new house without a foundation. Sure, it might have the greatest and most efficient new appliances and cool fixtures, but it’s not really going to work unless the fundamentals that support the entire structure are in place.

Too often, agencies begin their cloud migration efforts only to find out -- too late -- that their networks were unprepared for the transition. There have been tales of agency IT administrators running into bandwidth, security and network visibility challenges.

Administrators can avoid this pitfall by building modern networks designed for both the cloud of today and the needs of tomorrow. These networks must be software-defined, intelligent, open and able to accommodate both legacy technologies and more-advanced solutions during the cloud migration period. They must also be able to automatically scale for increased usage, potential security risks and future demands. Simultaneously, their administrators must have complete visibility into network operations and applications, wherever they may be hosted.

Let’s look at some building practices that administrators can use to effectively create a solid, modern and cloud-ready network foundation.

Create a blueprint to monitor bandwidth

Many network challenges will likely come from increased traffic derived from an onslaught of devices. The bring-your-own-device car drove out of the garage a long time ago and is now permanently parked in government agencies -- right next to issues related to the connected devices that make up the internet of things.

The result is that both traditional and non-traditional devices (and the data hogging, cloud-based applications they allow users to access) are enabling network traffic that will inevitably impact bandwidth. Backhaul issues can also occur, particularly with traditional network architectures that aren’t equipped to handle the load that more devices and applications can put on the network.

It’s becoming increasingly important for administrators to be able to closely monitor and analyze network traffic patterns. They must have a means to track bandwidth usage down to individual users, applications and devices so they can more easily pinpoint the root cause of slowdowns. The tools used to achieve both of these objectives must be open and completely customizable so the network itself can easily be adjusted and evolve to meet future demands.

Construct automated security protocols

Agencies moving from a traditional network infrastructure to the cloud will want to make sure their security protocols evolve as well. Administrators will need tools that can police their networks and immediately alert administrators to potential threats. Network software should automatically detect and report on potentially malicious activity, use of rogue or unauthorized devices and other factors that can prove increasingly hazardous as agencies commence their cloud migration efforts.

Indeed, automation will become vitally important to securing the cloud infrastructure. There are simply too many moving parts to a modern, cloud-ready network for managers to easily and manually control. In addition to the aforementioned monitoring practices, regular software updates should be automatically downloaded to ensure that the latest versions of network tools are installed. Patches should be automatically applied to correct potential security vulnerabilities. And administrators should consider instituting self-healing protocols that allow the network to automatically correct itself in case of a slowdown or breach.

Create an open-concept environment

Lack of visibility can be a huge network management challenge when migrating to the cloud. When everything was hosted on-site, visibility wasn’t a problem; everything was in the data center and easily accessible by administrators. That’s no longer necessarily the case, as government agencies are hosting more and more data in the cloud.

Agency IT personnel must be able to maintain a holistic view of everything that’s happening on the network, wherever that activity may be taking place. For those taking a hybrid cloud approach -- where some data remains on-site, while some is hosted in the cloud -- that will require network monitoring that allows them to see into the dark hallways that exist between on-premises and cloud infrastructures. They must be able to continuously monitor the performance of those applications, regardless of where they exist.

In many cases, the foundation that government networks are built upon now includes stacks of applications, each one highly dependent on the other. Like bricks in the wall of a house, they interlock to form a strong, complete whole that could be compromised if one of the bricks is missing or damaged. It’s critical that managers can see everything transpiring within this stack and be able to quickly identify and rectify potential issues that could impact other applications.

Much as well-built real estate increases in value over time, creating a cloud-ready, modernized network will offer significant benefits, both now and in the future. Agencies will be able to enjoy better security and greater flexibility through networks that can grow along with demand, and they’ll have a much easier time managing the move to the cloud with an appropriate network infrastructure. In short, they’ll have a solid foundation upon which to build their cloud strategies.

About the Author

Joe Kim is executive vice president engineering and global CTO at SolarWinds.

Stay Connected

Sign up for our newsletter.

I agree to this site's Privacy Policy.