public inbox for [email protected]  

help / color / mirror / Atom feed

(no subject)
2+ messages / 1 participants
[nested] [flat]

* (no subject)
@ 2019-05-19 18:43  Vanil Mar <[email protected]>
  0 siblings, 1 reply; 2+ messages in thread

From: Vanil Mar @ 2019-05-19 18:43 UTC (permalink / raw)
  To: [email protected]

Hi,



There was a project done over Postgresql 9.1.1

 My task now is to migrate only the modified optimizer and develop it
further on 10.4


How can I efficiently approach such a simulator? I mean I will need only to
provide specific trees and return results, not whole postgresql dbms. In
general, when I was developing over the whole postgresql it was fine as I
can create a db in the regular dbms and then just compile the modified
files and start it again over the already created database. Now I am really
not sure what should be the right and smart approach.



Can you share any advice/information/links/articles, please?

Thank you in advance!


^ permalink  raw  reply  [nested|flat] 2+ messages in thread

* Simulator; only modified postgresql optimizer
@ 2019-05-21 18:47  Vanil Mar <[email protected]>
  parent: Vanil Mar <[email protected]>
  0 siblings, 0 replies; 2+ messages in thread

From: Vanil Mar @ 2019-05-21 18:47 UTC (permalink / raw)
  To: [email protected]

Hi,



There was a project done over Postgresql 9.1.1 - it is a preference aware
query optimizer. It consists of modified parser(which provides the users
with an option for additional parser syntax) and modified optimizer(early
prunning, additional cost metric). It is generating optimized plan.


 My task now is to migrate only the modified optimizer and develop it
further on 10.4. This means that I will have to input directly a parser
tree to the optimizer module and output the new tree generated by the
optimizer.


How can I efficiently approach such a simulator? I mean, I will need only
to provide specific trees to the optimizer module and return results, not
to  develop a whole new version of postgresql dbms. In general, when I was
developing over the whole postgresql it was fine as I can create a db in
the regular dbms and then just compile the modified files and start it
again over the already created database. Now I am really not sure what
should be the right and smart approach.



Can you share any advice/information/links/articles, please?


>


^ permalink  raw  reply  [nested|flat] 2+ messages in thread

end of thread, other threads:[~2019-05-21 18:47 UTC | newest]

Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2019-05-19 18:43  Vanil Mar <[email protected]>
2019-05-21 18:47 ` Simulator; only modified postgresql optimizer Vanil Mar <[email protected]>

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox