3

I've been noticing that PgAdmin III isn't as responsive as I expected it to be.

For instance a query to select * ... from a table containing 20,397 rows takes 19.531 seconds in PgAdmin III and 7.383 seconds in DbVisualizer 9.0.7. I initially thought this was a coincidence, but it has been consistent across all queries run.

As such it appears that PgAdmin III takes approximately twice as long to query my database over DbVisualizer. So how can I speed up PgAdmin III Queries?

I am using Heroku Postgres 9.1.9 on the Crane package and PgAdmin 1.16.0.

--EDIT--

I upgraded PgAdmin to version 1.16.1 and query time is 19.038 seconds, so no change with software revision.

Improve this question
5
  • You didn't think the version of pgAdmin might be of interest?
    – Erwin Brandstetter
    Jun 24, 2013 at 23:06
  • I posted the version. Please see edit.
    – Robert H
    Jun 25, 2013 at 12:02
  • Have you tried to compare the time needed with that of psql? I think this difference is a result of displaying the results, over which the user has no real influence...
    – dezso
    Jun 26, 2013 at 6:07
  • @dezso psql gives me ERROR: character 0xefbfbd of encoding "UTF8" has no equivalent in "WIN1252" so unfortunately I don't have data from psql. However your displaying comment makes sense. I will do some additional testing on a linux machine to see if there is any difference.
    – Robert H
    Jun 26, 2013 at 12:57
  • @dezso psql run via Ubuntu 12.04 with \timing shows execution time of 1895.867 ms, where pgAdmin on the same machine takes 21502 ms. Looks like the client is the issue. If you add this as an answer I'll gladly accept. Thanks :-)
    – Robert H
    Jun 26, 2013 at 13:17

1 Answer 1

Reset to default
3

It looks like you see a difference between the clients you use.

In my experience the different PostgreSQL clients need very different times for displaying the data they got from the server. Far the fastest was psql and then pgAdminIII, EMS SQL Manager and phppgadmin in who-remembers-what order.

When tried with commands which return no or few rows, the difference is usually negligible or at least very small. Furthermore, you should see the same actual time consumed (with the usual small differences) when you issue an EXPLAIN ANALYZE ... command instead of the actual query which returns the data much slower in one client than the other.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.