Author: gabe appleton 2015-05-12 19:24:20
Published on: 2015-05-12T19:24:20+00:00
In a mailing list discussion amongst Bitcoin developers in May 2015, Gregory Maxwell expressed frustration at the lack of attention paid to desirable characteristics for block coverage. These included locality, compactness, computational efficiency, uniformity, and resistance to attacks. Maxwell proposed schemes that came close to meeting these criteria but failed on one or more fronts. One such scheme involved reservoir sampling and communicated only 64 bits of data to know what blocks a node claimed to have, but required O(height) work to figure out what blocks a peer had from the data communicated. Another scheme based on consistent hashes had log(height) computation but sometimes resulted in a node needing to go refetch an old block range it previously didn't store, creating re-balancing traffic. Maxwell believes it is possible to meet all of the desired criteria, but much time has not been spent on it.
Updated on: 2023-06-09T20:52:17.283995+00:00