Author: gladoscc 2015-11-10 01:58:41
Published on: 2015-11-10T01:58:41+00:00
A proposal was made to compress block datastreams before sending them to reduce bandwidth usage. The proposal suggested using Zlib compression for blocks, which resulted in an average of 20% compression and took 90 milliseconds to compress a full block. The savings would be mostly in terms of reducing bandwidth usage, but there could also be a small performance gain during the transmission of the blocks, particularly where network latency is higher. There was some debate over the value of the proposed 25% bandwidth savings, with some considering it considerable, especially for people running full nodes in countries like Australia where internet bandwidth is lower and there are data caps. Others were less convinced and felt that the implementation complexity tradeoff may not be worth it. Bob McElrath calculated compression ratios for zlib, gzip, and bzip2 compression, finding that zlib compression was right out, but gzip and bzip2 compression showed savings of 28% and 25%, respectively. A suggestion was made to give the BIP a more generic title, "Support for Datastream Compression," rather than the specific PR title of "Zlib Compression for block relay," since it could also be used for transactions at a later time.
Updated on: 2023-06-11T00:59:57.796525+00:00