Optimizing Public Sector IT Environments

Government IT shops have been told for years that they need to do more with less.

But “less” can only go so far. Eventually, agencies need more money to do new, innovative things that help advance their missions. And if they can’t get this funding through appropriations, they need to find ways to free up existing funds that aren’t serving them well. 

One obvious opportunity is to simply make existing environments, such as Amazon Web Services implementations, more efficient. Many agencies’ computing environments are the cumulative result of consultants coming and going over several years, and they have never been optimized for cost or performance.

Update cloud infrastructure: Public cloud vendors like AWS update their infrastructure every 12 to 18 months. But many customers let their cloud instances continue to run on older, less efficient CPUs. By temporarily shutting down a virtual machine in the cloud and switching it over to the latest generation of CPU — a process that takes just a few minutes — agencies can make their compute instances up to 10% more efficient. 

Optimize CPU architecture: This is a bit more involved than the previous step, but it also can yield outsized savings. Changing the CPU architecture of cloud instances often requires agencies to also modify their applications. When they do, their savings typically jump out of the single digits, reaching between 20% and 30%. Especially for agencies with large cloud environments, that can quickly add up to significant cost savings.

Eliminate dark data: Dark data is information that no one in the organization owns. Because no one is responsible for it, dark data typically sits and grows, taking up more and more resources while adding nothing of value. In fact, this data is often given VIP treatment, sitting on the most expensive storage equipment, because no one has an incentive to catalog it and move it to an archival storage tier. By simply making sure all significant data stores have an owner, agencies can dramatically reduce their storage bills.

Modernize code: Many software systems were built as temporary minimum viable products. Then, when the software did the job adequately, it was left in place permanently, with no real thought to efficient execution. With modern coding tools, including AI, agencies can take a fresh look at their applications and seek out opportunities for quick wins. This doesn’t necessarily mean rewriting everything from scratch. Rather, IT leaders should prioritize the fixes that will provide the greatest return for the effort.

Look to the edge: Agencies deploy IT resources at the network edge to put data and processing power closer to the users who need it, which helps reduce latency. But much of this infrastructure was put in place when wireless networks were slower and less reliable, and it might no longer be needed. Odds are, agencies can move much of these workloads to larger, more efficient clusters, either in a public cloud environment like AWS or in their own data centers.

The first step in optimizing any IT environment is an audit, often led by a trusted third-party partner like CDW, who can bring a fresh set of eyes to your infrastructure. After an audit, the “out of sight, out of mind” problem disappears, and IT leaders can implement better practices that improve security and efficiency, freeing up funds for new, mission-critical projects.