We can have pruned nodes that can verify blocks and transactions so the block size can be increased without leading to the centralization of block verification.

But if we increase block size then block verification time and propagation delay will increase. It will increase the probability of two miners mining a block simultaneously and also delay the competing blockchain branches to resolve the dispute. Probably any increase in block propagation delay will increase the time to resolve disputes exponentially. Based on a website (Can't remember) currently, block propagation delay is 5 seconds to reach 90% nodes which may increase to 15-20 seconds if all transactions were Segwit.

I think block propagation delay is the only limiting factor in increasing block size.

Has anyone done an investigation to find out the relationship between block propagation delay and optimal block size?

  • I think you may be thinking of dsn.kastel.kit.edu/bitcoin/#propagation – Murch May 18 at 12:59
  • Pruned nodes must still download and validate every block. So while concerns about storage space requirement are alleviated via pruning, concerns about required bandwidth and computation are not. – chytrik May 18 at 17:21
  • @murch.. yeah.. i got propagation delay from that website only.. i am thinking if we could somehow know the optimal block size without affecting Bitcoin's security then we can increase blocksize to that number... at this time network fee is fine but in future we may need bigger blocks.. such information will be useful.. – Vizeet Srivastava May 19 at 3:24
  • The website does not give average branching resolution time as factor of block size.. – Vizeet Srivastava May 19 at 3:29
  • I was just trying to point out where you might have found your propagation delay figure. – Murch May 21 at 18:56

I think block propagation delay is the only limiting factor in increasing block size.

This assumption is not accurate. While block propagation time is an important parameter for fair competition among miners, the block size also affects initial synchronization times of new nodes, disk space for archival nodes, and bandwidth (especially upload) requirements for all node operators.

Block propagation between miners should preferably take less than a second, but I don't think the block size and propagation are as related as the question surmises and thus consider the main question to be unanswerable.

  • Agreed. It was my mistake to not consider initial synchronization times as an important factor. – Vizeet Srivastava May 22 at 13:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.