Firms Need to Cut Hidden but Unnecessary IT Budget

As the COVID-19 has taken a toll on the global economy, the IT budget forecasts have slipped downwards; it’s time to tighten up the technical stack — before the CFO comes calling

In the current crisis, as the revenue evaporates, it’s time for firms to cut back by slicing out those extra expenditures and bold ideas that made sense in normal circumstances.

Virtual Meetings have become an Integral Part of the ‘New Normal’

Dump the gimmicks

Clever bits of data and flashy website displays are expected in good times — and can even increase revenue a pinch. But in lean times, they’re the easy target to save money, especially as such “enhancements” are usually supported by a distinct microservice running in its individual pod.

Change architectural priorities

Development teams usually try to meet their targets, emphasizing speed, such as response time. Saving those few milliseconds means adding tiers of servers and building elaborate networks closer to the users. But, now every penny matters, customers can do with less.

As the priority is switched from speed to efficiency, such extra layers of caching and synchronization have melted away. Sometimes slowing down by just 10 to 20 percent can save over half of the computational effort. Saving money on the extra resources helps to save the labor for keeping all of those layers running.

Audit infrastructure allocations

In tight times, dialing back the CPUs is usually a bit easier as the layers that allocate the cores are largely automated. If there’s no extra CPU core available, the software waits another nanosecond until one of them is clear.

Dialing back memory is a bit more dangerous because it’s common for software to crash or fail when it can’t find more memory. If the code fails gracefully, IT managers can watch the log files as the RAM gets reduced.

Rethink disaster preparations

It is strange to use a huge societal disaster as an excuse to go easy on disaster preparations, but it has proved to be the need of the hour now. Building a fail-safe, robust database for collecting orders for mission-critical healthcare materials has become more essential than ever.

But extending the same belts-and-suspenders principles to a bunch of social media posts is not. Some databases don’t need replication around the world every few milliseconds. Some keystrokes need not be tracked. Some databases don’t require transaction consistency. Many bits of data don’t need attention at all. A set of log files for some transactions is sufficient for bits that are referenced occasionally, if at all.

Switch to serverless

Over the past few years, a new option for “serverless” has gained popularity, which still involves server hidden under layers. This helps to simplify computing because the infrastructure takes care of starting and shutting down virtual servers when the occasional request comes up.

The prices are usually mind-bogglingly small as cloud vendors charge the smallest fractions of a cent for responding to most of the web request. With databases or websites that only welcome limited people a month, the bill could be less than a penny, and it might go down to zero as well.

Can Machine Learning Compensate for Lateral Thinking?

Search out the servers with the least load and look to replace them. They might be experimental tools or the ones for business niches. Firms need to be careful with those that might encounter a usage spike or go viral. The fractions of a penny add up quickly when too many start showing up.

These small areas need greater attention in the current pandemic situation to ensure that optimizing the tight budget.