This is not a new article or even a new subject – the matter of the ‘zettabyte era’ has been the subject of a few articles from as far back as 2015. I first came across the concept of the Zettabyte Era from this article from Real Clear Energy as part of their Energy and the Information Infrastructure series. I thought it might be worth revisiting the article and consider the implications for the ordinary companies who want to get on with the business of managing customers and orders but find themselves in the world of an increasingly complex world of data.
So, the obvious question is what is a zettabyte? It’s a number so big, it defies simple explanation, there no avoiding an analogy or comparison to get your head around it… I can’t do better than this article did the first time around…
“For those not steeped in numerical prefixes, each name represents a 1,000-fold jump: terabyte (trillion), petabyte (thousand trillion), extabyte (thousand petabytes), and then the zettabyte. A zetta stack of dollar bills would go from the earth to the sun -- that’s 93 million miles away -- and back, 700 thousand times. Such scales defy imagination.”
Or if you prefer a visualisation of a zettabyte, compared to the paltry number that is a million (or even a billion for the Musk’s and Bezos’ amongst you)..
The 1000 fold increase in each name change conceals the understanding of the scale to the point where we know that a billionaire is a rich, but they are 1000 times richer than a millionaire. Common sense, but….. Anyway, we’ve labored the point of how big a zettabyte is - big - but what does it mean in the real world?
Data is invisible, we don’t see it flying through the air like a string of 1’s and 0’s towards the WiFi router and as such we forget it’s there, like oxygen, keeping our daily lives breathing. Oxygen, however, is in limited supply, data clearly is not. Unless we start to plant more trees, and a lot, and allow them to mature into the ancient woodlands we as a species decimated in just a few short decades we will never have as much oxygen on planet earth as we have had in the past.
Data, however, is growing at a logarithmic scale and it’s a parallel worth exploring.
Data = Energy. In simplistic terms, you can’t process data without expending some energy. Even writing a note on paper uses a pen, paper, some calorific content to fuel your brain and your muscles. The same goes for your laptop, your mobile phone, your iPad and the massive cloud resources that power the ability of your hand-held mobile device to place bunny ears on your selfies and videos. We consume these resources with few thoughts as to the cost, financial or otherwise. Let’s look at it another way. Your typical laptop consumes about 100w of power and that can be translated into BTUs.
BTU - The British thermal unit (BTU or Btu) is a unit of heat; it is defined as the amount of heat required to raise the temperature of one pound of water by one degree Fahrenheit
BTUs give us the ability to compare in basic terms the comparative energy parallels of abstract things. For example, the amount of energy in a tree and the amount of energy used by computers. Cleary there’s lots of different laptops, people prefer their screens more or less bright and trees come in many varieties of soft and hard wood which affects how much energy they can provide, but for the sake of illustration we’re going to use a typical home laptop and oak as our wood of choice.
Some quick maths.
Timber is typically measured in a quantity called a ‘cord’. A cord of wood is a pile of logs 4 feet wide, 4 feet high and 8 feet long.
Burning a cord of seasoned oak produces roughly 27 million BTUs but a good proportion of that is lost in the generation and supply chain and this useful diagram from www.enerdynamics.com illustrates that well.
So, as a guide 70% of energy is lost before it reaches your laptop charger, meaning our cord of wood has produced 9 million BTUs.
Your lap top consumes 100 watts of electricity per hour and I’m going to assume you use it for 8 hours per day and a watt hour is equivalent to 3.41 BTU’s so…
So our modest store of wood above would keep your laptop running for about 9 years. But that’s just the laptop not the cloud services like Netflix that are being consumed while you are using it.
I must admit, I didn’t believe this number when I first calculated it. So I went back to basics and to the ‘Primary Generation Efficiency’ in the diagram above. Another way to calculate the electricity output of our cord of wood would be to consider what is known as the ‘Heat Rate’ of the generation plant. That is to say, how efficient is the biomass being burnt and the plant burning it at generating watts of power. As ever, there’s a range and the larger wood burning plants may achieve a 10,000 heat rate whilst new small more efficient heat capturing plants can achieve 20,000 het rate.
So, revisiting the calculations from first principles we get this.
Almost exactly the same number. But I’m new to the intricacies of BTUs and Kilowatt Hours so I decided to base the calculation on an an actual power station with published numbers, the Drax Wood Pellet Powered Station in the UK. There’s still a few assumptions in here, such as how much weight of ‘raw wood’ is needed to create dry wood pellets, but its indicative none the less.
What I wanted to do here was just make sure the calculations were realistic in the sense that they weren’t totally at the wrong end of the spectrum, (that is to say a cord of wood actually runs your laptop for 60 minutes!). So we have something to think about: -
A cord of wood will run a laptop for about 10 years or run 10 laptops for a year.
The Drax power station above will power 51 million laptops for a year.
These numbers are starting to make some sense: -
Computer ownership (not including computers in workplace or educational settings) is around 60% per capita for the UK (60% of every adult and child) that gives around 40 million laptop or desktop computers in UK households
The UKs largest ‘renewable source’ (we’ll come back to that…) power station therefore produces enough energy to run everybody’s computers for the equivalent of 15 months.
Just the computer attached to your charger. Not the services they use like Netflix.
Not lighting, heating, ovens, fridges, industrial manufacturing, street lights, kettles, showers, hospitals, TVs, smart phones, iPads, cloud data centres, Twitter, Facebook or Instagram… Just your laptop, plugged in, not communicating with anything and with its user not consuming any other energy like making a cup of tea.
Processing data uses energy.
The Drax power station above creates around 15 Terawatt/hours of electricity per year.
The UKs consumption of electricity in 2021 was 334 Terawatt/hours
In this context, the phrase “Data is the new oil” has a different meaning. The world is still addicted to oil and there no sign of that addiction waning anytime soon. The graph below taken from OPECs World Oil Outlook is illustrative of that.
The world’s new addiction to data is no less important than the headline grabbing consumption of oil. A big proportion of data processing will be driven by fossil fuel energy sources. Even renewable energy sources and sustainable energy sources are sometimes not what they seem. In the context of the zettabyte era, Data is the New Oil, but not in the positive way in which that term was originally coined.
So what next for data traffic prediction? The zettabyte milestone was passed a few years ago, and we head inexorably onwards. The Internet of Things still has to live up to its full growth and potential, but as we have more cars, fridges, doorbells and lightbulbs attached to the internet the volume of traffic heads towards Yottabytes and Brontobytes of data. A Brontobyte, a million times bigger than a Zettabyte.
As children, we were told to switch off the lights when we leave the room. As adults we’re told to not to use the car for short journeys. With data usage and energy consumption growing at such a phenomenal rate we need new mantras to live by. We’ll explore how we can think differently about data in a later article. For now, I’ll leave you with this thought and the ramifications it has for data growth in the future..
A new type of chip can transmit all of the internets traffic every second