Re: [PoC] Improve dead tuple storage for lazy vacuum

On Mon, Nov 14, 2022 at 3:44 PM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
wrote:
>
> 0004 patch is a new patch supporting a pointer tagging of the node
> kind. Also, it introduces rt_node_ptr we discussed so that internal
> functions use it rather than having two arguments for encoded and
> decoded pointers. With this intermediate patch, the DSA support patch
> became more readable and understandable. Probably we can make it
> smaller further if we move the change of separating the control object
> from radix_tree to the main patch (0002). The patch still needs to be
> polished but I'd like to check if this idea is worthwhile. If we agree
> on this direction, this patch will be merged into the main radix tree
> implementation patch.

Thanks for the new patch set. I've taken a very brief look at 0004 and I
think the broad outlines are okay. As you say it needs polish, but before
going further, I'd like to do some experiments of my own as I mentioned
earlier:

- See how much performance we actually gain from tagging the node kind.
- Try additional size classes while keeping the node kinds to only four.
- Optimize node128 insert.
- Try templating out the differences between local and shared memory. With
local memory, the node-pointer struct would be a union, for example.
Templating would also reduce branches and re-simplify some internal APIs,
but it's likely that would also make the TID store and/or vacuum more
complex, because at least some external functions would be duplicated.

I'll set the patch to "waiting on author", but in this case the author is
me.

--
John Naylor
EDB: http://www.enterprisedb.com