For all the noise tech gurus make about artificial intelligence, it’s still frustratingly elusive.
Google, the granddaddy of them all recently dropped the signal on their BIG DATA SEARCH network as their centralized enterprise servers browned-out for several hours. YouTube, Shopify, and SnapChat, all tied to the sputtering Google Cloud, went dark. According to reports, no one on Google’s tech team were able to prevent the downtime for the company and their customers–even though they knew it was happening.
How does something like this occur, which btw, is not the first time for Google or other enterprise cloud companies like Amazon. Google’s recent fail however was the biggest event of this type in history. Networks get knocked out every day, but we don’t often hear about it because it doesn’t always directly impact the public. This time, it did.
Is it the fault of the tech team? Well kinda-sorta. Mostly though the problem is in the chain of command, with ultimate responsibility in the hands of the c-suite oracle at the top — usually, either the CEO or CTO, plus often a board of directors and advisors … and not to be forgotten of course, pesky shareholders.
For all the noise companies also make about flat, lateral-management styles, claiming that everyone on the team has an equal say, all I can say is …hmm. When you look closely, it turns out there are many versions of flat organizations, and each adaptation is unique to each company.
Under these conditions, it’s a little foggy and hard for some tech teams to plot a failsafe course, and because of this, many techs shoulder overwhelming responsibility, but conversely also feel like they don’t share the accolades and benefits.
When I heard about the Google brown-out, like millions of others I couldn’t help but wince. Incidents like this cause systemic distrust in our industry, which has slowly grown over the decades to where today we no longer really trust tech in the manner we did a generation ago. Arguably, society still reveres IT talking heads like Gates, Bezos, and Musk, but it’s the tech soldiers on the front line who actually develop the solutions that make things happen. IPFS didn’t write itself.
IT teams often know about impending critical issues in their networks — usually well before anyone else in the company. However, if they don’t have “effective” communication skills and don’t explain the situation in simple terms to help c-suiters understand the urgent need for a timely response, the problem could spontaneously combust and hit critical overload in a nano-blip.
This latest tech glitch turned into a global Google Cloud embarrassment that cost millions.
One would think that with all the advances made in AI, a simple network glitch could maybe be automatically managed, but it can’t, which means we need better networks and human intervention.
If a small predictable glitch can knock out a heavyweight like Google, how hard do you think it is to take down a city’s government-run power or transportation grids? The recent epidemic of chain-supply hacking is a perfect example of how easy it is to disrupt a centralized network.
We now know for certain that many centralized enterprise networks are often so patchwork-spindly and back-door heavy you don’t need much to bring them down. Way too many legacy networks are STILL anemically cobbled together on a house of HTTP cards, which means disaster is only a huff and puff away.
Jump forward five years to a time when, according to today’s tech gurus, we’ll have self-driving vehicle and drones zipping all over our neighborhoods. What do you think will happen when someone hacks centralized highway networks and simultaneously turns all the traffic lights green at every intersection in your city? Tech claims it could never happen … I’m not so sure.
What happens too when the healthcare grid that supports your hospital also gets hacked?
Is there a solution? Yes, and it’s relatively easy.
Simply put, it’s called Decentralizing Big Data.
Decentralizing displaces the power from the monopolies, oligopolies, and single sources who control it now, and shares it securely and equally across distributed grids of independent servers called nodes.
That’s basically it … we all know what’s wrong, but unfortunately, like life, knowing isn’t enough.
Currently, the bigger problem is communication, not technology.
IT teams need to have confidence in their expertise so they can explain themselves more clearly. If your critical enterprise still runs on HTTP, well, you’re actually part of the problem. IT teams have the option to share their knowledge now, or suffer the consequences when the dam bursts, but remember, it never really does a career any good to say, “I told you so.”
It’s very likely your CEO or CTO is busy managing a long list of just-as-challenging issues, which means they need you to help them understand the ramifications of running outdated technology.
The BIGGEST FEAR reported by CEOs and CTOs is to be perceived as incompetent. Consequently, in frustration they quite often default to SAFE mode, which means they need more information than you might think.
The solution is knowledge, and effectively sharing that knowledge with people who make the decisions–the gatekeepers, and sometimes even with stakeholders who have incredible influence over CEOs.
The public is losing faith — not in tech, but in the people responsible for it, so change the game.
Artificial intelligence can solve a lot of problems, but it doesn’t have the type of intelligence necessary to make decisions that first require a responsible human decision. AI can’t solve complicated big data issues on a broken network, no matter how smart it is.
If you’re on the front line but you don’t know exactly how to communicate up-the-chain, contact us. We can point you in the right direction, which also might help you figure out your next career move.
In part 2 of this series I’ll look at specific solutions like Temporal for this ongoing BIG DATA challenge.