The CryptoLions history node is the subject of a lot of internal discussion.
They are not our biggest expense — our robust block producing node costs more, for example — but they are a significant expense.
We deployed a Mongo DB history node when Block.One announced that their History Plugin was being depreciated. We also made a Mongo DB history API which very closely copied the depreciated History Plugin: https://github.com/CryptoLions/EOS-mongo-history-APIT
We liked Mongo because if offered many new possibilities beyond the old history plugin’s capabilities. For example, by adding filters, you could search by actions instead of just getting actions by accounts. Our Mongo History node and plugin seemed to fill a niche at a critical time. Many history nodes had not yet transitioned from the old History Plugin and Dapps needed a quick, ready solution.
Right now, it is at 3.5 TB, and peaks at over 50 requests per second at peak hours, which isn’t as high as it used to be.
It seems a lot of Dapps are migrating to Elastic Search (EOS Tribe), DFuse (EOS Canada), or to block.one’s new State History Plugin which has cool new features like fork detection. Things are changing remarkably fast, as seems to be the standard for the EOS community.
Our node is not going anywhere yet. It’ll continue to operate, but we’ve been watching the evolution of history nodes, and deciding what to do next.
CryptoLions is an EOS Block Producer based in Ukraine. We strive to make EOS more valuable by building projects that improve the ecosystem, by setting the standard for transparency and accountability, and by popularizing EOS all over the world.