PgCache in the press — articles, podcasts, and talks from around the web.
James Nelson, PgCache's CTO, talks with Command Prompt founder Joshua Drake about PostgreSQL's conservative approach to development, forks, and pluggable storage. The episode ends with the problem that led to PgCache: files served at the edge while authorization checks still went back to a central database.
Play the episode →James Nelson's PostgreSQL wishlist: schema change events in logical replication. A cache kept current from the replication stream gets no event when a table is dropped, so nothing tells it to invalidate the cached results.
Play the episode →An independent benchmark by Hulunlante Worku, testing PgCache 0.6.2 against a 16-million-row dataset: 2.9s to 0.6ms on a 10M-row join, and no measurable difference on point lookups. Warm cache against warm cache, with every filter and join column indexed first.
Read at dev.to →A talk by PgCache's founders, James Nelson and Philip Johnston: why they built PgCache, how it uses logical replication to keep cached results current, and what they learned along the way.
Watch the talk →How PgCache uses the logical replication stream as a change feed to keep cached query results synchronized with the origin database — no manual cache invalidation logic required.
Read at The Read Replica →A wire-compatible PostgreSQL proxy that caches read queries and maintains freshness through logical replication, caching only frequently accessed data and updating results automatically when the underlying data changes.
Read at I Programmer →