Wednesday, July 08, 2015

What is the minimum energy cost of a blockchain?

Concern exists that securing bitcoin's blockchain is wasting energy. Begs the question - what is the information-theoretic minimum amount of energy needed to secure a globally distributed decentralized ledger with at least 1Mb blocks and the security parameters of the current blockchain?

I'm not even sure how to articulate the question, or frame it in an appropriate way. A mathematician could help here.

Its been suggested that a more useful proof of work that hashing to a difficulty-prescribed number of leading zeros would make better use of the energy invested. My intuition tells me that there is no free lunch - that if an appropriate puzzle were found (appropriate meaning necessary blockchain parameters met, for example, block approximately every 10 minutes, etc...) there would be additional energy required to make up for the impedance mismatch of aligning the nature of the useful problem solved to the requirements of securing the blockchain. But I surely lack the math rigor to back the intuition.

Another suggestion is to rely on some resource that is already wasted; excess disk space for example. Not sure.

Perhaps Landauer's Principle may have some utility here:
https://en.wikipedia.org/wiki/Landauer's_principle

Your thoughts?

No comments: