Re: Proposal : Parallel Merge Join

On Fri, Feb 24, 2017 at 3:04 PM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
> What advantage do you see for considering such a path when the cost of
> innerpath is more than cheapest_total_inner? Remember the more paths
> we try to consider, the more time we spend in the planner. By any
> chance are you able to generate any query where it will give benefit
> by considering costlier innerpath?

Changed
>
> 2.
> +static void generate_parallel_mergejoin_paths(PlannerInfo *root,
> + RelOptInfo *joinrel,
> + RelOptInfo *innerrel,
> + Path *outerpath,
> + JoinType jointype,
> + JoinPathExtraData *extra,
> + Path *inner_cheapest_total,
> + List *merge_pathkeys);
>
> It is better to name this function as
> generate_partial_mergejoin_paths() as we are generating only partial
> paths in this function and accordingly change the comment on top of
> the function. I see that you might be naming it based on
> consider_parallel_*, however, I think it is better to use partial in
> the name as that is what we are doing inside that function. Also, I
> think this function has removed/changed some handling related to
> unique outer and full joins, so it is better to mention that in the
> function comments, something like "unlike above function, this
> function doesn't expect to handle join types JOIN_UNIQUE_OUTER or
> JOIN_FULL" and add Assert for both of those types.

Done
>
> 3.
> A test case is still missing.
Added

--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com