Percona 對於 PostgreSQL 使用 HugePages 的評論

開頭我先說一下我的想法,我對於 Percona 的 Ibrar Ahmed 的文章保持著懷疑的態度,因為他先前在「Benchmark PostgreSQL With Linux HugePages」這篇做的 benchmark 就有奇怪的結果,但卻給不出合理的原因,甚至連 Percona 自家的 CEO 公開在 comment 問之後也沒有看到文章提出合理的解釋:


A lot of interesting results here…

1) PgBench access distribution is very interesting. With database size growing by 20% from 80G to 96G we see performance drop of Several times which is very counter-intuitive

2) There is no difference between 2MB and 4K but huge difference between 1G and 2M even though I would expect at least some TLB miss reduction in the first transitioning. I would understand it in case transparent huge pages are Enabled… but not disabled

3) For 96GB why would throughput grow with number of clients for 1G but fall for 2M and 4KB.

這次看到「Settling the Myth of Transparent HugePages for Databases」這篇,也是在討論 Linux 的 HugePages 對 PostgreSQL 帶來的影響,同樣馬上又看到奇怪的東西...


Figure 1.1 PostgreSQL’ s Benchmark, 10 minutes execution time where database workload(48GB) < shared_buffer (64GB)

Figure 1.2 PostgreSQL’ s Benchmark, 10 minutes execution time where database workload (48GB) > shared_buffer (64GB)

不過這邊可以推測 Figure 1.2 應該是 112GB (因為對應的圖片上面標的是 112GB),當做是標錯就好。

但這樣又跑出一個奇怪的結果,48GB 的資料量比較小,TPS 大約是 35K/33K/41K,但 112GB 資料量比較大,卻可以達到 39K/43K/41K~42K,反而比較快?我暫時想不到什麼理由...

整體的測試有 pgbench 與 sysbench (這邊也打錯成 sysbecnch,先不管),其中 pgbench 跑了 10 mins 與 60 mins 的版本,但是 sysbench 只跑了 10 mins 的版本?這是什麼原因...

另外還是有些情況是打開 HugePages 比較快的 (sysbench 的 64 clients),如果以直覺來說的話,我反而還是會打開 HugePages (yeah 純粹是直覺),我現在比較想知道他會在 Percona 裡面待多久...

PostgreSQL 對 fsync() 的修正

上次寫了「PostgreSQL 對 fsync() 的行為傷腦筋...」提到 fsync() 有些地方是與開發者預期不同的問題,但後面忘記跟進度...

剛剛看到 Percona 的人寫了「PostgreSQL fsync Failure Fixed – Minor Versions Released Feb 14, 2019」這篇才發現在 2/14 就出了對應的更新,從 release notes 也可以看到:

By default, panic instead of retrying after fsync() failure, to avoid possible data corruption (Craig Ringer, Thomas Munro)

Some popular operating systems discard kernel data buffers when unable to write them out, reporting this as fsync() failure. If we reissue the fsync() request it will succeed, but in fact the data has been lost, so continuing risks database corruption. By raising a panic condition instead, we can replay from WAL, which may contain the only remaining copy of the data in such a situation. While this is surely ugly and inefficient, there are few alternatives, and fortunately the case happens very rarely.

A new server parameter data_sync_retry has been added to control this; if you are certain that your kernel does not discard dirty data buffers in such scenarios, you can set data_sync_retry to on to restore the old behavior.

現在的 workaround 是遇到 fsync() 失敗時為了避免 data corruption,會直接 panic 讓整個 PostgreSQL 從 WAL replay 記錄,也代表 HA 機制 (如果有設計的話) 有機會因為這個原因被觸發...

不過也另外設計了 data_sync_retry,讓 PostgreSQL 的管理者可以硬把這個 panic 行為關掉,改讓 PostgreSQL 重新試著 fsync(),這應該是在之後 kernel 有修改時會用到...

從 Microsoft SQL Server 轉移到 PostgreSQL 的工具

在「How to Migrate from Microsoft SQL Server to PostgreSQL」這邊看到作者的客戶需要把 Microsoft SQL Server 轉移到 PostgreSQL (但沒有提到原因)。

裡面主要是兩個階段的轉換,第一個階段是 schema 的轉換,作者提到了 dalibo/sqlserver2pgsql 這個用 Perl 寫的工具:

Migration tool to convert a Microsoft SQL Server Database into a PostgreSQL database, as automatically as possible

第二個階段是資料的轉換,是選擇用 Pentaho Data Integration 的 Community Edition:

Pentaho offers various stable data-​centric products. Pentaho Data Integration (PDI) is an ETL tool which provides great support for migrating data between different databases without manual intervention. The community edition of PDI is good enough to perform our task here. It needs to establish a connection to both the source and destination databases. Then it will do the rest of work on migrating data from SQL server to Postgres database by executing a PDI job.

所以用兩個工具串起來... 另外在文章裡面沒提到 stored procedure 之類的問題,應該是他們的客戶沒用到或是很少用到?

PostgreSQL 對 fsync() 的行為傷腦筋...

FOSDEM 2019 上的演講,討論 PostgreSQL 在確保 ACID 特性中的 Durability 時遇到 fsync() 的行為跟預想的不一樣 (主要是當 fsync() 失敗的行為):「PostgreSQL vs. fsync」。

在「PostgreSQL vs. fsync. How is it possible that PostgreSQL used fsync incorrectly for 20 years, and what we'll do about it.」這邊的 Q&A 形式的訪談有快速描述了短期的計畫與長期的想法:

The short-term solution is ensuring that we detect fsync errors reliably at least on sufficiently recent kernels (since 4.13). On older kernels we can’t do much better, unfortunately.

The long-term solution is still being discussed in the community, but it’s hard to say how we could keep relying on buffered I/O in the future. So we may end up with direct I/O, but that’s a pretty significant change and is likely going to be a multi-year project.

MySQL 這邊則是以 O_DIRECT 為主的世界,受到的影響就小很多了...

Braintree (PayPal) 用 PostgreSQL 的方式

RDBMS 最困難的事情都圍繞在「怎麼不中斷服務」(很多事情在不用考慮 uptime/downtime 的前提下很好做,不論是 ALTER 或是 failover,到備份還原計畫),而 PayPalBraintree 在「PostgreSQL at Scale: Database Schema Changes Without Downtime」這邊討論修改 PostgreSQL 的 database schema 時怎麼不中斷服務。

文章內的大部份都是給 DBA 知道的細節 (e.g. 怎麼樣才不會觸發大規模的 lock 導致服務中斷),而不是開發者面向的事情... 但開頭的部份,也是我認為最重要的部份,則是需要 Developer 參與的:

For all code and database changes, we require that:

  • Live code and schemas be forward-compatible with updated code and schemas: this allows us to roll out deploys gradually across a fleet of application servers and database clusters.
  • New code and schemas be backward-compatible with live code and schemas: this allows us to roll back any change to the previous version in the event of unexpected errors.

為了符合這兩個要素,可能會在 schema 設計上有好幾個階段的操作,而非一次到位。而且也才能避免要關站從 backup 倒資料回來的情況...


大家在猜 Amazon DocumentDB 的底層是不是 PostgreSQL...

Amazon DocumentDB 的出現讓人驚訝的倒不是 AWS 推出這塊服務,而是 AWS 對於這類對 PaaS 有攻擊性的 license model 的反擊姿態。這也導致了在 AWS 推出後 MongoDB 的股價掉了 13%。

另外一方面,大家也都想要知道 AWS 怎麼堆底層的系統,畢竟要從頭開發一個所需要的功夫應該不小... (雖然 AWS 應該有這個能力)

從「Is DocumentDB really PostgreSQL?」這邊看到 Hacker News 上的「My bet is that it is built on top of Aurora PostgreSQL.」這篇討論,透過目前 DocumentDB 的限制,大家在猜 Amazon DocumentDB 是不是拿 PostgreSQL 改出來的...

目前看起來 Identifiers 的 63 chars 限制,單一 collection 的 32TB 限制 (對應到表格),以及 UTF-8 null character 限制,都跟 PostgreSQL 一樣。

也許過一陣子 AWS 的人會找個地方透漏,不過目前看起來只能猜...

PostgreSQL 的 Amazon RDS Database Preview Environment

AWS 推出了 PostgreSQL 試爆場讓大家測試 XDDD:「Amazon RDS Database Preview Environment is now available」。

可以測還沒有正式 release 的版本:

The Amazon RDS Database Preview Environment is now available, offering an environment for customers to easily test beta, release candidate, and early production versions of PostgreSQL database engine software with the convenience and flexibility of Amazon RDS.


Percona 宣佈提供 PostgreSQL Support...

有點意外的,Percona 宣佈提供 PostgreSQL Support:「Percona Expands Services Offerings with PostgreSQL Support」。看起來是建出熟 PostgreSQL 的團隊來做生意了...

之後應該會在他們站上開始看到與 PostgreSQL 相關的文章了;另外一方面,對於付錢買 Percona 服務的人來說,可以跟同一個廠商通包所有的服務 (從 MySQLMariaDBPercona Server 這三個 MySQL 系列的資料庫,到 MongoDB,然後是昨天推出的 PostgreSQL):

As a result, organizations can, for the first time, work with a single trusted vendor to meet their support needs for MySQL, MongoDB, MariaDB, PostgreSQL, or any hybrid combination of these database technologies, whether deployed on-premises, in the cloud, or in a Database as a Service (DBaaS) environment.

然後也許有機會看到 Percona 客製的 PostgreSQL?

Amazon RDS 宣佈支援 PostgreSQL 10

Amazon RDS 宣佈支援 PostgreSQL 10 了:「PostgreSQL 10 now Supported in Amazon RDS」。而且 AWS 這次推出的還包括了 10.1 的 patch:

As of version 10, PostgreSQL no longer uses three-part version numbers, and is shifting to two-part version numbers. This release includes all patches from the PostgreSQL 10.1 minor version.

10 的第一個版本是去年十月初 (在「PostgreSQL 10 Released」這邊可以看到),10.1 是去年十一月初 (在「PostgreSQL 10.1, 9.6.6, 9.5.10, 9.4.15, 9.3.20, and 9.2.24 released!」),現在二月底,所以延遲大約是三個多月的時間...

10.2 是二月初,不知道會多久...

用 Percona Monitoring and Management (PMM) 蒐集 PostgreSQL 的數據

難得在 Percona 的 blog 上看到專門談 PostgreSQL 的文章:「Collect PostgreSQL Metrics with Percona Monitoring and Management (PMM)」。

其實是透過 Prometheus 疊出來的:

Starting from PMM 1.4.0. it’s possible to add monitoring for any service supported by Prometheus.


3. In the next dialog, choose Prometheus as a data source and continue.