Synchronous streaming replication/Standby nodes in a cluster for an HA solution
The post presents a proposal to implement an HA solution based on PostgreSQL streaming replication and Standby node structure. This solution is still in construction, so the final implementation design...
View ArticlePostgreSQL: playing with foreign data wrappers (1)
This post presents some basics when using foreign data wrappers with PostgreSQL for external files. FOREIGN DATA WRAPPER is a part of SQL/MED (Management of external data with SQL) and its...
View ArticleTake care of kernel memory limitation for postgresql shared buffers
When tuning a PostgreSQL server, one the major setting parameters is the one controlling the amount of shared memory allowed with shared_buffers. PostgreSQL has a default shared_buffers value at 32MB,...
View ArticleUnlogged table performance in postgresql 9.1
This study is made with PostgreSQL 9.1.1, released a couple of days before this post is written. Unlogged tables are a new performance feature of PostgreSQL 9.1, created by Robert Hass. So, by guessing...
View ArticleCollation in PostgreSQL 9.1
Collation is a new functionality of PostgreSQL 9.1 that allows to specify the sort order by table column or for an operation. In Ubuntu, the collation types supported by your system are listed in...
View ArticleCode evolution between Postgres 9.1 and 9.2
For the last couple of days I have been working on merging the code of Postgres-XC from 9.1 to 9.2. The release 1.0 of XC has been based on 9.1, but it is time to move forward and steal the latest...
View ArticlePostgres feature highlight: hstore
hstore is a PostgreSQL contrib module in core code for a pretty long time. Its code is located in contrib/hstore in source folder. It is particularly useful to store sets of key/value in a single table...
View ArticlePostgres feature highlight: pg_buffercache
pg_buffercache is a PostgreSQL contrib module allowing to get an instant relation-based view of the shared buffer usage by querying the wanted server. This can be pretty useful for performance analysis...
View ArticlePostgres FDW highlight: redis_fdw
A foreign-data wrapper (FWD) in a Postgres server allows to fetch data from an foreign entity or a foreign server. In this case, the Postgres planner and executer have the notion of what is called a...
View ArticlePostgres 9.3 feature highlight: timeline switch of slave node without archives
Since PostgreSQL 9.1, it is possible to switch a standby server to follow another server that has been freshly promoted after the master node of a cluster is out due to a failure (disaster or another)....
View Article
