Re: [PATCH] Incremental sort (was: PoC: Partial sort) - Mailing list pgsql-hackers

From Justin Pryzby
Subject Re: [PATCH] Incremental sort (was: PoC: Partial sort)
Date
Msg-id 20200407042521.GH2228@telsasoft.com
Whole thread Raw
In response to Re: [PATCH] Incremental sort (was: PoC: Partial sort)  (Tomas Vondra <tomas.vondra@2ndquadrant.com>)
Responses Re: [PATCH] Incremental sort (was: PoC: Partial sort)
Re: [PATCH] Incremental sort (was: PoC: Partial sort)
List pgsql-hackers
On Mon, Apr 06, 2020 at 09:57:22PM +0200, Tomas Vondra wrote:
> I've pushed the fist part of this patch series - I've reorganized it a

I scanned through this again post-commit.  Find attached some suggestions.

Shouldn't non-text explain output always show both disk *and* mem, including
zeros ?

Should "Pre-sorted Groups:" be on a separate line ?
| Full-sort Groups: 1 Sort Method: quicksort Memory: avg=28kB peak=28kB Pre-sorted Groups: 1 Sort Method: quicksort
Memory:avg=30kB peak=30kB
 

And, should it use two spaces before "Sort Method", "Memory" and "Pre-sorted
Groups"?  I think you should maybe do that instead of the "semicolon
separator".  I think "two spaces" makes sense, since the units are different,
similar to hash buckets and normal sort node.  

     "Buckets: %d  Batches: %d  Memory Usage: %ldkB\n",
     appendStringInfo(es->str, "Sort Method: %s  %s: %ldkB\n",

Note, I made a similar comment regarding two spaces for explain(WAL) here:
https://www.postgresql.org/message-id/20200402054120.GC14618%40telsasoft.com

And Peter E seemed to dislike that, here:
https://www.postgresql.org/message-id/ef8c966f-e50a-c583-7b1e-85de6f4ca0d3%402ndquadrant.com

Also, you're showing:
    ExplainPropertyInteger("Maximum Sort Space Used", "kB",
                groupInfo->maxMemorySpaceUsed, es);

But in show_hash_info() and show_hashagg_info(), and in your own text output,
that's called "Peak":
        ExplainPropertyInteger("Peak Memory Usage", "kB", memPeakKb, es);
    ExplainPropertyInteger("Peak Memory Usage", "kB",
                spacePeakKb, es);

-- 
Justin

Attachment

pgsql-hackers by date:

Previous
From: Noah Misch
Date:
Subject: Re: Spurious "apparent wraparound" via SimpleLruTruncate() rounding
Next
From: Amit Kapila
Date:
Subject: Re: doc review for parallel vacuum