PGConf US 2017 was held in Jersey City from March 28 to 31. Every year, PGConf US provides collaboration and updates on PostgreSQL, the most advanced open source database in the world. Bitnine participated and presented AgensGraph.
Bitnine attended many events, which will be presented below. Also, an AgensGraph announcement will be made at the end of this blog.
Logical Replication in PostgreSQL 10 [link]
A presentation from a developer on the logical replication function built into PostgreSQL 10. This feature can be used as an extension called logical. Logical replication can do partial replication which is impossible with existing physical replication methods sending WAL logs. It can also be used for data migration when upgrading PostgreSQL versions.
Jsonb in Postgres – the present and future [link]
A presentation by the core developers about the history and current status of JSON support in PostgreSQL, and future plans. We were able to hear the order of “hstore → json → nested hstore → jsonb” and the introduction of SQL / JSON, the standard for JSON processing in SQL, and the implementation status of SQL / JSON in PostgreSQL.
The support function is on the level of review. This will be available in PostgreSQL 11. Also, we were able to see the comparison of performance and efficiency with other databases with compression jsonb.
All the dirt on VACUUM: 9.6 edition [link]
The presentation was on why using the vacuum is necessary implementing mvcc in PostgreSQL. This presentation explains why vacuum should be used, what happens if you use vacuum, and what is the bloating of the index.
Next-Generation Parallel Query [link]
A new parallel query feature in PostgreSQL 9.6 supports parallel processing only for sequential scans among multiple scan methods. In this release, we discussed enhanced parallel query functionality, including index scans added to PostgreSQL 10 to be released soon, and parallel functionality for bitmap heap scans, and shared the performance test results and future issues to be addressed.
What is in PostgreSQL 10.0 [link]
A new function in PostgreSQL 10 that will be coming soon. Logical replication, parallel queries, and the new authentication methods have been the main focus.
Moving PostgreSQL Forward [link]
Presentation on the current status of the PostgreSQL community. They shared a summary of how many people contribute to PostgreSQL development and how much one contributes. We were also able to see what the PostgreSQL community is doing in order to grow and the things to consider that are preventing growth. Conclusively, it was obvious that we need a lot of people to contribute to the development.
A Practical Postgres Protocol Primer [link]
A brief presentation on the frontend / backend protocol currently used in PostgreSQL. They first explained the message sent and received during query execution and the certification course in order with sample code. And they explained how to cancel query execution.
Implementing Graph Database based on PostgreSQL [link]
In this presentation, we introduced AgensGraph, a graph database based on PostgreSQL, and shared development issues. The presentation was conducted by Dr. Kim Ki-sung, CTO of Bitnine. Despite the final announcement of the “Internals” track on the last day, many people attended.
Starting from company introduction, Bitnine began to explain what the graph database was, what data models AgensGraph selected, and the superiority of AgensGraph by comparing it with some graph database performance
There was a Q & A session after the announcement and it was a good opportunity for many people to ask questions and confirm their interest in the graph database.
Although it was a short two-day event, it was a good opportunity to see people’s interest in PostgreSQL and to meet several companies. Also, it was a good time to inform the world of the AgensGraph and to share opinions about the graph database industry. All of the announcements have been recorded and will soon be shared on the web. Please refer to PGConf 2017 for more details.
BITNINE GLOBAL INC., THE COMPANY SPECIALIZING IN GRAPH DATABASE
비트나인, 그래프 데이터베이스 전문 기업