Aggregate (cost=8581.22..8581.23 rows=1 width=4) (actual
time=1209.498..1210.530 rows=1 loops=1)
Buffers: shared hit=3108 read=4314 dirtied=1
I/O Timings: shared read=1169.977
-> Gather (cost=8580.80..8581.21 rows=4 width=8) (actual
time=1209.371..1210.526 rows=5 loops=1)
Workers Planned: 4
Workers Launched: 4
Buffers: shared hit=3108 read=4314 dirtied=1
I/O Timings: shared read=1169.977
-> Aggregate (cost=7580.80..7580.81 rows=1 width=8) (actual
time=1206.518..1206.520 rows=1 loops=5)
Buffers: shared hit=3105 read=4314 dirtied=1
I/O Timings: shared read=1169.977
-> Hash Join (cost=2620.70..7578.45 rows=937 width=0) (actual
time=148.965..1206.330 rows=1897 loops=5)
Buffers: shared hit=3105 read=4314 dirtied=1
I/O Timings: shared read=1169.977
-> Index Only Scan using
IX_Collaborators_TenantId_IsDeleted_RunningIndex on Collaborators c
(cost=0.55..4619.26 rows=11407 width=80) (actual time=2.815..1058.793 rows=7474
loops=5)
Index Cond: ((c."TenantId" =
'0f8026c775a5499fdd3ccd193a15f94c3b4ae58a'::text) AND (c."IsDeleted" = false))
Heap Fetches: 3711
Buffers: shared hit=2332 read=3563
I/O Timings: shared read=1028.548
-> Hash (cost=2479.01..2479.01 rows=11292 width=79)
(actual time=144.988..144.989 rows=1983 loops=5)
Buffers: shared hit=637 read=751 dirtied=1
I/O Timings: shared read=141.429
-> Index Only Scan using
IX_IntegrationCollaboration_TenantId_IsIntegrationDeleted_IsDe~ on
IntegrationCollaboration i (cost=0.55..2479.01 rows=11292 width=79) (actual
time=1.804..144.122 rows=1983 loops=5)
Index Cond: (i."TenantId" =
'0f8026c775a5499fdd3ccd193a15f94c3b4ae58a'::text)
Heap Fetches: 112
Buffers: shared hit=637 read=751 dirtied=1
I/O Timings: shared read=141.429
Planning:
Buffers: shared hit=440
Execution time: 1210.592 ms
So we can see here that the query read 4314 blocks in 1.2 sec.. (ive done 4314
* 8kb / 1.2sec which is about 28 MB/s)…
Hope you’ll have any clue 😊
From: Christophe Pettus <[email protected]>
Date: Monday, 31 March 2025 at 21:44
To: Eden Aharoni <[email protected]>
Cc: [email protected] <[email protected]>
Subject: Re: [EXTERNAL] RDS IO Read time
> On Mar 31, 2025, at 11:30, Eden Aharoni <[email protected]> wrote:
>
> Can you please tell me what other data might help?
You can show your work on how you got the megabytes/second number. (Be aware
that on a general open-source mailing list, there's only so much debugging that
we can do of a specific problem that is not a PostgreSQL bug.)
