Fresh Facts & Numbers :: Exact prediction of Blockchain size depending on the transaction volume.

It's true that Bitcoin will become more centralized over time. The capital costs of mining will grow. But not necessarily because of this issue. 100 terabytes today costs around $6000, and the price of that will fall over five years. Mining is not a huge a concern, though, because miners don't have to hold the entire blockchain. Other parts of the network will become more centralized. ~Benjamin Dees, http://www.reddit.com/user/benjamindees in the Reddit topic called Still don't understand how we gonna be able to handle Bitcoin if its Blockchain size will be over 100TB 

Mr.Dees came to this conclusion perhaps because there's widely distributed opinion about not being able to handle blockchain.

I write this post, to help Bitcoin users quit worrying about huge size of a Blockchain (Main Distributed Public Ledger of any cryptocurrency, especially Bitcoin - it is the file where all of the transactions stored since the Bitcoin creation).

We can know exact amount of transactions in Bitcoin network (since its creation in 2009), today (Jun 2014) the rough number of total transactions of Bitcoin is 40 millions (check this link to see updated numbers).

In order to find out how much Blockchain will be in size in few years from now, let's begin with transaction industry leaders facts:

• Global non-cash payment transactions already grew 8.8% to reach 307 billion transactions in 2011 - Global non-cash payments volumes are expected to top 333 billion transactions in 2012 after transactions grew by 8.8 percent in 2011, according to the latest available data from the World Payments Report 20131 published by Capgemini and RBS.

If 7 billion people on Earth would use non-cash payments, this report would mean that 1 person performs around 50 transactions a year. But perhaps not every of 7 billion people on our planet right now are using non-cash payments, that's why it wouldn't be correct to hope that we will overcome this average number of transactions (333 billions, maybe few hundred billions more). So I'll use this number of transactions globally as basis to technical analysis for estimating Bitcoin's Blockchain size in future.

If you'll consider that Bitcoin will totally takeover that "non-cash" market it will means that the yearly volume of transactions (quantity of transactions) will be around 333 billions.

What it will mean for the Blockchain?

In order to calculate, how much megabytes of information every transaction will take in a blockchain, in case if transaction volume will start growing exponentially and will achieve the numbers around 333 billions which we already have in "non-cash" payments, I've examined how much data each transaction usually adds up to the Blockchain.

Approximately each 1 transaction for Bitcoin means around 575 bytes added to the size of Blockchain (Note that for Litecoin we have different average numbers - in Litecoin 1 transaction averagely takes around 1257 bytes - which is almost 1KB - so Litecoin's Blockchain theoretically should grow 2 times faster than Bitcoin's due to its scrypt architecture and its frequency of the blocks confirmed into the Blockchain).

Let's just imagine that next year all of the "non-cash" transactions become Bitcoin transaction, and here we have it: THE 333 BILLION TRANSACTIONS PER YEAR in BITCOINS!

So, it basically means: 27,75 billions transactions per month TRANSFERRED IN BITCOINS!

To compare with current amount of transactions in Bitcoin, it is: 40 million transactions! Tiny amount, in comparison with current global non-cash transactions quantity.

And I will remind you with 40 million transactions, our blockchain size is 23 GIGABYTES (up to Jun 2014)

Still you don't think that blockchain size isn't a problem for Bitcoin? Then, you just have to make some basic math: Average transaction size as I mentioned before is 575 bytes - that is about half of a kilobyte - that's how much disk space takes 1 transaction.

Now it should be easy for you to find out that roughly 27 billions per month (or 333blns/year) will take 27000000000 * 575 = 15525000000000 (or 15 terrabytes, 525 gigabytes).

I'll repeat.

Still you don't think that blockchain size isn't a problem for Bitcoin?

If my calculations is correct, then maybe you might just don't believe in such exponential growth for Bitcoin.

But if with non-cash payments - we understand, that the most population on the Earth use non-cash, just because they need to receive their salary, then do some few purchases/transfers which is possible only through non-cash transactions. And the common numbers showed us that average amount of transactions on each person on the earth is 50. But what if the rest of the Earth will get more active? Double this number, and you will get 100 transactions per each world's citizen, and that will mean double growth for the blockchain size each month - not 333 billions per year, but 666 per year, and not 15 terrabytes per month, but 30 terrabytes per month instead.

What if total Bitcoin will replace the paper cash? (and again that's adds up a lot more transactions to the so-called "non-cash" transactions worldwide).

What is the solutions?

Solution #1. Prune the blockchain, or introduce the new version of blockchain every month.

I think this is a great solution, because it is very easy to check whether old version blockchain balances corresponds to the balances transferred into a new version of blockchain. But in this case we will get another central point of failure: software managing company like "Bitcoin Foundation" which will presumably play role in introducing those versions of blockchain will have the authority to stop doing that in case if the regulators will find the ways to approach the offices of the Bitcoin Foundation / or arrest the owners of the software company.

Probably the whole process of switching from one version of blockchain to another could be decentralized by software solution, but I think that's sounds like another experiment which could even break the world's economy if some failure will happen during the experiment (Imagine that the whole system is broken at the volume of 333 billions of transactions)

So in this case my solution is pruning Blockchain each time it achieves 30 terrabytes or some other reasonable quantity of data. We can link those blockchains into the other structure - the global blockchain of blockchains :) Sounds funny, but as I understand this might be the case if the quantity of transactions will grow crazily.

Solution #2. The other solutions.

• Each country creates its own cryptocurrency. Takes some marketing and cryptographic initiative from governments/corporations
• The blockchain gonna become segmented and used only for large transactions while locally all the other transactions will be transacted through the third parties / other decentralized entities. I'm not studied this way deeply enough. 

So we will definitely have fun by creating those solutions