Author: Dave Scotese 2015-08-09 20:43:48
Published on: 2015-08-09T20:43:48+00:00
On August 9, 2015, a discussion occurred on the bitcoin-dev mailing list regarding the issue of the software not coping well with a continuously growing mempool. As a permanent solution, Thomas Zander proposed a user-configurable default limit to the size of the mempool regardless of block size. The idea was to protect those who run full nodes from bitcoind eating more and more memory. In response, Dave Scotese argued that transaction backlog has valuable solutions other than increasing the block size. He suggested enabling senders to increase their fee when necessary and prioritizing this over increasing the transactions per second that the network can handle. The competition between these two is unfair because of how easy it is to apply the "N MB-blocks bandaid." Regarding the argument that backlogs in the past have already trained users to wait for at least one confirmation or go off-chain, Scotese disagreed and stated that he doesn't think he's all that different from others, and that is what he has done. While recognizing the difficulty of double-spending, he also did some research on replace by fee (FSS-RBF) and on Child-pays-for-parent (CPFP) and pointed out that these solutions are "not supported in the vast majority..." Scotese concluded that the trend seems to be towards improvements and that transaction backlog has valuable solutions other than increasing the block size. In summary, while Zander proposed a permanent solution to protect full nodes from bitcoind eating more and more memory, Scotese argued that there are other valuable solutions to transaction backlog, such as enabling senders to increase their fee when necessary, and prioritizing this over increasing the transactions per second that the network can handle.
Updated on: 2023-06-10T18:31:50.117374+00:00