Near future

  • Lazy value type cache, per-conn binary I/O decision.

  • Clean up the binary-or-not decision. There should be possible to have partitions with different versions. Requires lazy values.

Good to have

  • RUN ON ALL: ignore errors?

  • RUN ON ANY: if one con failed, try another

Just thoughts

  • Drop plproxy.get_cluster_config()…

  • Streaming for big resultsets, to avoid loading them fully in memory. This needs also backend changes.

  • integrate with memcache:

    set_object(id, data)
    CACHE SET object(id) = data;
    get_object(id)
    CACHE GET object(id)