Re: slow index scan performance

[Soni M <diptatapa@xxxxxxxxx>]

Trying on another server, it gives different result.

->  Index Scan using response_log_by_activity on public.response_log rl2  (cost=0.00..50.29 rows=17 width=8) (actual time=0.955..0.967 rows=0 loops=30895)

                                       Output: rl2.activity_id, rl2.feed_id

                                       Index Cond: (rl2.activity_id = rl.activity_id)

                                       Filter: rl2.success

                                       Buffers: shared hit=2311312 read=132342

                           ->  Index Scan using activity_pkey on public.activity a  (cost=0.00..49.79 rows=1 width=12) (actual time=13.747..13.762 rows=1 loops=30892)

                                 Output: a.status_id, a.activity_id, a.visit_id

                                 Index Cond: (a.activity_id = rl.activity_id)

                                 Buffers: shared hit=124463 read=30175

Now, index scan on activity_pkey which take much slower. Can someone please explain these ?

Thanks

On Tue, Sep 5, 2017 at 8:46 PM, Soni M <diptatapa@xxxxxxxxx> wrote:

It's Postgres 9.1.24 on RHEL 6.5

On Tue, Sep 5, 2017 at 8:24 PM, Soni M <diptatapa@xxxxxxxxx> wrote:

Consider these 2 index scan produced by a query

->  Index Scan using response_log_by_activity on public.response_log rl2  (cost=0.00..51.53 rows=21 width=8) (actual time=9.017..9.056 rows=0 loops=34098)

                                       Output: rl2.activity_id, rl2.feed_id

                                       Index Cond: (rl2.activity_id = rl.activity_id)

                                       Filter: rl2.success

                                       Buffers: shared hit=3357159 read=153313

                           ->  Index Scan using activity_pkey on public.activity a  (cost=0.00..51.10 rows=1 width=12) (actual time=0.126..0.127 rows=1 loops=34088)

                                 Output: a.status_id, a.activity_id, a.visit_id

                                 Index Cond: (a.activity_id = rl.activity_id)

                                 Buffers: shared hit=137925 read=32728

And it's size

conscopy=# select pg_size_pretty(pg_relation_size('response_log_by_activity'::regclass));

 pg_size_pretty 

----------------

 7345 MB

(1 row)

conscopy=# select pg_size_pretty(pg_relation_size('activity_pkey'::regclass));

 pg_size_pretty 

----------------

 8110 MB

(1 row)

Index scan on response_log_by_activity is far slower. The table has just been repacked, and index rebuilt, but still slow.

Is there any other way to make it faster ?

Why Buffers: shared hit=3,357,159 read=153,313 on response_log_by_activity is much bigger than Buffers: shared hit=137925 read=32728 on activity_pkey while activity_pkey size is bigger ?

--

Regards,

Soni Maula Harriz

--

Regards,

Soni Maula Harriz

--

Regards,

Soni Maula Harriz