Sysbench tests on PG12

Started by Subhrajyoti Senapatiover 4 years ago3 messagesgeneral
Jump to latest
#1Subhrajyoti Senapati
ssubhrajyoti@google.com

Hi,

Was running a few sysbench tests in Postgres12.
Sysbench Test Config
oltp-readwrite-custom
Threads - 500
Machine: 16 core 64G

In PG server:
shared_buffers: 16GB
maintenance_work_memory: 16GB
checkpoint_timeout: 1h
max_wal_size: 10GB
wal_sync_method: 'open_sync'
effective_cache_size: 32GB
wal_buffers: -1
wal_compression: on

[image: line-graph (6).png]
Got some qps dips.

Changed fsync and full_page_rewrite to 'off'.
Which helped in avoiding these many long dips in performance:

[image: line-graph (9).png]
But fsync ensures DBMS consistency. So, instead of that can I change a few
other configurations to avoid these inconsistent performance dips ?

Regards,
Subhrajyoti

Attachments:

line-graph (6).pngimage/png; name="line-graph (6).png"Download+2-2
line-graph (9).pngimage/png; name="line-graph (9).png"Download
#2Laurenz Albe
laurenz.albe@cybertec.at
In reply to: Subhrajyoti Senapati (#1)
Re: Sysbench tests on PG12

On Fri, 2021-10-01 at 14:22 +0530, Subhrajyoti Senapati wrote:

Was running a few sysbench tests in Postgres12.
Sysbench Test Config
oltp-readwrite-custom
Threads - 500
Machine: 16 core 64G

In PG server:
shared_buffers: 16GB
maintenance_work_memory: 16GB
checkpoint_timeout: 1h
max_wal_size: 10GB
wal_sync_method: 'open_sync'
effective_cache_size: 32GB
wal_buffers: -1
wal_compression: on

Got some qps dips.

Changed fsync and full_page_rewrite to 'off'.
Which helped in avoiding these many long dips in performance:

But fsync ensures DBMS consistency. So, instead of that can I change a few other configurations to avoid these inconsistent performance dips ?

Try playing with "bgwriter_flush_after", "checkpoint_flush_after" and "wal_writer_flush_after"
and see if that influences the behavior.

Yours,
Laurenz Albe
--
Cybertec | https://www.cybertec-postgresql.com

#3Subhrajyoti Senapati
ssubhrajyoti@google.com
In reply to: Laurenz Albe (#2)
Re: Sysbench tests on PG12

Thanks Laurenz, will try these flags.

Regards,
Subhrajyoti

On Fri, Oct 1, 2021 at 5:19 PM Laurenz Albe <laurenz.albe@cybertec.at>
wrote:

Show quoted text

On Fri, 2021-10-01 at 14:22 +0530, Subhrajyoti Senapati wrote:

Was running a few sysbench tests in Postgres12.
Sysbench Test Config
oltp-readwrite-custom
Threads - 500
Machine: 16 core 64G

In PG server:
shared_buffers: 16GB
maintenance_work_memory: 16GB
checkpoint_timeout: 1h
max_wal_size: 10GB
wal_sync_method: 'open_sync'
effective_cache_size: 32GB
wal_buffers: -1
wal_compression: on

Got some qps dips.

Changed fsync and full_page_rewrite to 'off'.
Which helped in avoiding these many long dips in performance:

But fsync ensures DBMS consistency. So, instead of that can I change a

few other configurations to avoid these inconsistent performance dips ?

Try playing with "bgwriter_flush_after", "checkpoint_flush_after" and
"wal_writer_flush_after"
and see if that influences the behavior.

Yours,
Laurenz Albe
--
Cybertec | https://www.cybertec-postgresql.com