RE: [Patch] Optimize dropping of relation buffers using dlist

On Thursday, November 12, 2020 1:14 PM, Tsunakawa-san wrote:
> The patch looks OK.  I think as Thomas-san suggested, we can remove the
> modification to smgrnblocks() and don't care wheter the size is cached or not.
> But I think the current patch is good too, so I'd like to leave it up to a
> committer to decide which to choose.
> I measured performance in a different angle -- the time
> DropRelFileNodeBuffers() and DropRelFileNodeAllBuffers() took.  That
> reveals the direct improvement and degradation.
> 
> I used 1,000 tables, each of which is 1 MB.  I used shared_buffers = 128 MB
> for the case where the traditional full buffer scan is done, and shared_buffers
> = 100 GB for the case where the optimization path takes effect.
> 
> The results are almost good as follows:
> 
> A. UNPATCHED
> 
> 128 MB shared_buffers
> 1. VACUUM = 0.04 seconds
> 2. TRUNCATE = 0.04 seconds
> 
> 100 GB shared_buffers
> 3. VACUUM = 69.4 seconds
> 4. TRUNCATE = 69.1 seconds
> 
> 
> B. PATCHED
> 
> 128 MB shared_buffers (full scan)
> 5. VACUUM = 0.04 seconds
> 6. TRUNCATE = 0.07 seconds
> 
> 100 GB shared_buffers (optimized path)
> 7. VACUUM = 0.02 seconds
> 8. TRUNCATE = 0.08 seconds
> 
> 
> So, I'd like to mark this as ready for committer.
I forgot to reply.
Thank you very much Tsunakawa-san for testing and to everyone
who has provided their reviews and insights as well.

Now thinking about smgrnblocks(), currently Thomas Munro is also working on implementing a 
shared SmgrRelation [1] to store sizes. However, since that is still under development and the
discussion is still ongoing, I hope we can first commit these set of patches here as these are already
in committable form. I think it's alright to accept the early improvements implemented in this thread
to the source code.

[1] https://www.postgresql.org/message-id/CA%2BhUKG%2B7Ok26MHiFWVEiAy2UMgHkrCieycQ1eFdA%3Dt2JTfUgwA%40mail.gmail.com

Regards,
Kirk Jamison