I wrote a blog post recently about my favourite podcasts. One of those I listed was Background Briefing, a radio program broadcast by the Australian Broadcasting Corporation (Australia’s ABC). A recent episode entitled Fatigue Factor really sparked my interest. It talked about the affects of fatigue on professions such as:
- Air Traffic Controllers
- Train drivers
- Truck drivers
It contained some alarming facts about the potential affects of fatigue and is well worth taking the time to listen to. However in my opinion there was one major omission:
It did not mention workers in the IT industry.
For many years I worked as the Account Engineer for several of IBM’s System z customers, mainly banks. Most weekends I skipped Saturday night as a sleep night. If I was lucky I might get to sleep from 10pm to 1am and would then head off to vast, noisy, dehydrating air conditioned computer rooms to perform various system changes. If I did my job well, had no hardware issues and the client confirmed everything was running as expected, I got to head home about 7am on Sunday. So that night I would have slept somewhere between zero and three hours. I would then spend the rest of the week recovering, before doing it all over again the following weekend at a different customer.
I mention all of this because fatigue was something I learnt to live with. Even when I moved to a support role, I still occasionally worked through the night on critical situations (something IBM calls Crit Sits). I also worked on a support roster which could involve 3am callouts to assist my fellow IBMers across the Asia Pacific region. So when I later moved to a Pre-Sales role, it certainly did wonders in helping me re-establish normal sleep patterns.
Listening to this podcast really brought home to me that the IT industry is just as guilty of failing to deal with fatigue, as the other industries that the podcast discusses. Now if your thinking this means it’s an IBM problem, think again. Most weekends I was working alongside representatives from EMC, HDS, Storagetek, etc. Plus of course there were the clients themselves, many of whom were also missing a nights sleep to satisfy their change and business requirements.
One of the major issues raised in the podcast is that there is no accepted way to measure how fatigued an employee actually is. This is a major problem. There are established tests to confirm how affected someone is by alcohol, or by drugs. But we cannot easily confirm how badly fatigued a worker is; plus many people are unwilling or unable to admit that they are suffering from fatigue
If we think about many of the major IT related outages that have occurred recently, I ponder what role fatigue played in each one. Even if it didn’t cause the initial issue, did making your employees work around the clock to resolve an issue, actually extend the outage time? For example, have a read of Amazons explanation of its recent Service Disruption. Just picking on some of the lines in the report:
At 12:47 AM PDT on April 21st, a network change was performed…
At 2:40 AM PDT on April 21st, the team deployed a change…
By 5:30 AM PDT, error rates and latencies again increased ….
At 11:30AM PDT, the team developed a way to prevent….
Was the person doing the change working out of their usual sleep pattern? Was the team working to resolve the issue working out of their normal sleep pattern? Did fatigue compound the outage? Its an interesting idea. Now it may well be that fatigue had NOTHING to do with this outage. It is pure speculation on my part. But I am certain that the root causes of many of the recent IT meltdowns and their extended after affects (such as Sony’s ongoing issues), MUST include the debilitating affects of fatigue.
Plus here is another rather disturbing fact. To quote from the podcast:
… if you’re sleep deprived, you’re more likely to crave chips over lettuce, and feel less like climbing the stairs. And that can become a vicious cycle, because many people who are overweight are even more prone to sleep disorders….