On 2022-Sep-20, Matthias van de Meent wrote: > Allow me to add: compressability > > In the thread surrounding [0] there were complaints about the size of > catalogs, and specifically the template database. Significant parts of > that (688kB of 8080kB a fresh PG14 database) are in pg_rewrite, which > consists mostly of serialized Nodes. If we're going to replace our > current NodeToText infrastructure, we'd better know we can effectively > compress this data.
True. Currently, the largest ev_action values compress pretty well. I think if we wanted this to be more succint, we would have to invent some binary format -- perhaps something like Protocol Buffers: it'd be stored in the binary format in catalogs, but for output it would be converted into something easy to read (we already do this for pg_statistic_ext_data for example). We'd probably lose compressibility, but that'd be okay because the binary format would already remove most of the redundancy by nature. Do we want to go there? -- Álvaro Herrera 48°01'N 7°57'E — https://www.EnterpriseDB.com/ "Java is clearly an example of money oriented programming" (A. Stepanov)
