Обсуждение: [ADMIN] Why pgpool TPS is lowest versus postgresql direct connections?

Поиск
Список
Период
Сортировка

[ADMIN] Why pgpool TPS is lowest versus postgresql direct connections?

От
"Lazaro Garcia"
Дата:

The correct results reported by sysbench was:

 

Concurrent Users

1

20

50

100

PostgreSQL

3582

11943

12852

10618

Pgpool

2240

7628

7013

6135

 

Is there any way to tuning this behavior?

 

Regards

 

 

De: Lazaro Garcia [mailto:lazaro3487@gmail.com]
Enviado el: miércoles, 8 de febrero de 2017 05:28 p. m.
Para: 'pgpool-general@pgpool.net'; 'pgsql-admin@postgresql.org'
Asunto: Why pgpool TPS is lowest versus postgresql direct connections?

 

After installed Pgpool with 2 postgresql nodes with streaming replication, I have noticed that access directly to postgresql is more efficient than through pgpool.

 

I supposed that load balance could increase the transactions per second executed because each node could receive more load, but the results shown below are not expected.

 

This is the setup:

 

Pgpool 3.6.1 whit connection pooling, streaming replication mode and load balancing mode.

 

2 PostgreSQL server 9.6.1 whit streaming replication.

 

For the tests I used sysbench and pgbench.

 

The results of sysbench:

 

Concurrent Users

1

20

50

100

PostgreSQL (TPS) Direct

1166

20936

25743

27344

Pgpool (TPS)

2240

7628

7013

6135

 

 

The results of pgbench

 

1

20

50

100

PostgreSQL (TPS) Direct

1403

6805

6194

5726

Pgpool (TPS)

511

5430

5528

4705

 

 

As you can see in both cases even with load balance, the total transactions per second are lower.

 

Is this the expected behavior. Is there any way to allow more TPS when pgpool is used?

 

There are other publications with similar results:

 

https://www.os3.nl/_media/2011-2012/courses/lia/rory_breuk_gerrie_veerman_-_report.pdf    (page 28)

http://www.mail-archive.com/pgpool-general@pgfoundry.org/msg03326.html

 

 

Regards

[ADMIN] Re: [pgpool-general: 5318] Why pgpool TPS is lowest versus postgresql direct connections?

От
"David Sisk -X (dsisk - TEKSYSTEMS INC at Cisco)"
Дата:

My immediate thought would be this:  PGPool is much slower with a larger number of concurrent users because some of the connections are being queued by PGPool. Take a look at your num_init_children parameter…unlike most connection pooling layers, PGPool doesn’t have separate config parameters for the initial number of backend connections and the max number of backend connections…num_init_children controls both.  Increase num_init_children to at least the number of concurrent connections plus a few, and re-try the benchmark tests.

 

Also, the benefits of load balancing don’t tend to show up well with pgbench or sysbench default benchmarks.  PGPool imposes some overhead in determining which queries can be load-balanced against a standby…with SELECT queries that execute in only a few milliseconds, it might take PGPool longer to make that decision than it would have to just send it straight to the primary. So, you won’t see the load balancing benefits with SELECT queries that execute in a few milliseconds, but when you send it SELECT queries that take a few seconds or longer to execute you’ll start to see the benefits.  You could create a custom benchmark that has some high-reduction queries that would take a few seconds to execute, then run the benchmarks with that…you should then see scalability improvements from the load balancing.

 

Hope this helps,

 

banner2

 

David Sisk

Engineer - Software

dsisk@cisco.com

Tel:

Cisco Systems, Inc.

7025-6 Kit Creek Road PO Box 14987
RESEARCH TRIANGLE PARK
27709-4987
United States
cisco.com

 

http://www.cisco.com/assets/swa/img/thinkbeforeyouprint.gifThink before you print.

This email may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message.

Please click here for Company Registration Information.

 

From: pgpool-general-bounces@pgpool.net [mailto:pgpool-general-bounces@pgpool.net] On Behalf Of Lazaro Garcia
Sent: Thursday, February 9, 2017 5:00 AM
To: pgpool-general@pgpool.net; pgsql-admin@postgresql.org
Subject: [pgpool-general: 5318] Why pgpool TPS is lowest versus postgresql direct connections?

 

The correct results reported by sysbench was:

 

Concurrent Users

1

20

50

100

PostgreSQL

3582

11943

12852

10618

Pgpool

2240

7628

7013

6135

 

Is there any way to tuning this behavior?

 

Regards

 

 

De: Lazaro Garcia [mailto:lazaro3487@gmail.com]
Enviado el: miércoles, 8 de febrero de 2017 05:28 p. m.
Para: 'pgpool-general@pgpool.net'; 'pgsql-admin@postgresql.org'
Asunto: Why pgpool TPS is lowest versus postgresql direct connections?

 

After installed Pgpool with 2 postgresql nodes with streaming replication, I have noticed that access directly to postgresql is more efficient than through pgpool.

 

I supposed that load balance could increase the transactions per second executed because each node could receive more load, but the results shown below are not expected.

 

This is the setup:

 

Pgpool 3.6.1 whit connection pooling, streaming replication mode and load balancing mode.

 

2 PostgreSQL server 9.6.1 whit streaming replication.

 

For the tests I used sysbench and pgbench.

 

The results of sysbench:

 

Concurrent Users

1

20

50

100

PostgreSQL (TPS) Direct

1166

20936

25743

27344

Pgpool (TPS)

2240

7628

7013

6135

 

 

The results of pgbench

 

1

20

50

100

PostgreSQL (TPS) Direct

1403

6805

6194

5726

Pgpool (TPS)

511

5430

5528

4705

 

 

As you can see in both cases even with load balance, the total transactions per second are lower.

 

Is this the expected behavior. Is there any way to allow more TPS when pgpool is used?

 

There are other publications with similar results:

 

https://www.os3.nl/_media/2011-2012/courses/lia/rory_breuk_gerrie_veerman_-_report.pdf    (page 28)

http://www.mail-archive.com/pgpool-general@pgfoundry.org/msg03326.html

 

 

Regards

Вложения

Re: [ADMIN] Why pgpool TPS is lowest versus postgresql direct connections?

От
Jehan-Guillaume de Rorthais
Дата:

Hi, Le 9 février 2017 10:59:53 GMT+01:00, Lazaro Garcia <lazaro3487@gmail.com> a écrit :
[...]
>Is there any way to tuning this behavior?

The fact is that pgpool need to parse every single query going through it, guess if the query is a RO or a Rw, then
decidewhat to do with it. 

Which means:
- It adds a latency on every single query
- pgpool itself can become a bottleneck, especially during bench with lot of small and fast queries
- Pgpool handles N client connections and N*M backends connection
- It tracks xact, prepared queries, etc

In short, pgpool is not transparent and free of load...



>De: Lazaro Garcia [mailto:lazaro3487@gmail.com]
>Enviado el: miércoles, 8 de febrero de 2017 05:28 p. m.
>Para: 'pgpool-general@pgpool.net'; 'pgsql-admin@postgresql.org'
>Asunto: Why pgpool TPS is lowest versus postgresql direct connections?
>
>
>
>After installed Pgpool with 2 postgresql nodes with streaming
>replication, I
>have noticed that access directly to postgresql is more efficient than
>through pgpool.
>
>
>
>I supposed that load balance could increase the transactions per second
>executed because each node could receive more load, but the results
>shown
>below are not expected.
>
>
>
>This is the setup:
>
>
>
>Pgpool 3.6.1 whit connection pooling, streaming replication mode and
>load
>balancing mode.
>
>
>
>2 PostgreSQL server 9.6.1 whit streaming replication.
>
>
>
>For the tests I used sysbench and pgbench.
>
>
>
>The results of sysbench:
>
>
>
>
>Concurrent Users
>
>1
>
>20
>
>50
>
>100
>
>
>PostgreSQL (TPS) Direct
>
>1166
>
>20936
>
>25743
>
>27344
>
>
>Pgpool (TPS)
>
>2240
>
>7628
>
>7013
>
>6135
>
>
>
>
>
>The results of pgbench
>
>
>
>
>1
>
>20
>
>50
>
>100
>
>
>PostgreSQL (TPS) Direct
>
>1403
>
>6805
>
>6194
>
>5726
>
>
>Pgpool (TPS)
>
>511
>
>5430
>
>5528
>
>4705
>
>
>
>
>
>As you can see in both cases even with load balance, the total
>transactions
>per second are lower.
>
>
>
>Is this the expected behavior. Is there any way to allow more TPS when
>pgpool is used?
>
>
>
>There are other publications with similar results:
>
>
>
>https://www.os3.nl/_media/2011-2012/courses/lia/rory_breuk_gerrie_veerman_-_
>report.pdf    (page 28)
>
>http://www.mail-archive.com/pgpool-general@pgfoundry.org/msg03326.html
>
>
>
>
>
>Regards

--
Envoyé de mon appareil Android avec K-9 Mail. Veuillez excuser ma brièveté.