The previous “You Heard it Here First” post included a number of orders of magnitude approximations, especially in regards to zettabytes and the human connectome.
To be a bit more accurate, all the world’s digital data is actually estimated to be around 200 zettabytes (a fifth of a yottabyte!). That compares to “only” 1-2 zettabytes to capture and process a single human connectome. So all the world’s current data could actually capture a few connectomes, if we could figure out how.
As a further clarification, that capturing and processing of a connectome will require many orders of magnitude more storage than is ultimately needed to store the result. The same proved true for the human genome, which serves as an excellent, realistic example of things to come:
- A single human genome is about 6 billion base pairs, which ultimately only requires about one gigabyte to store (1/1000th of a terabyte). However the “raw data” generated to determine and confirm what each of those base pairs is takes about a petabyte (one million gigabytes) total.
- Similarly, a single human connectome is estimated to be about 100 trillion synapses, which would “only” take about a petabyte to store (a mere 1000 terabyte hard disks). But the raw data generated to characterize and map those synapses is where the needed storage gets to that estimated 1-2 zettabytes.
And hopefully that’s enough for now!