Linux kernel patches boost PG performance up to twice:
https://www.phoronix.com/news/MGLRU-Reaches-mm-stable
https://www.phoronix.com/news/MGLRU-Reaches-mm-stable
Phoronix
MGLRU Patches Merged To "mm-stable" Ahead Of Linux 6.1 - New Benchmarks Look Good
As further indication of MGLRU hopefully being mainlined for Linux 6.1 as planned, the Multi-Gen LRU patches have now been moved to Andrew Morton's mm-stable branch.
At what conference for the second year in a row was the author of this channel not invited to speak? :)
In short, about MERGE in just released Postgres 15:
https://twitter.com/samokhvalov/status/1580935490580058112
https://twitter.com/samokhvalov/status/1580935490580058112
Twitter
Two things MERGE lacks now:
- RETURNING clause to allow user to see what exactly was done and save on RTT
- WHEN NOT MATCHED BY [SOURCE|TARGET] to allow easy syncing with source - e.g., ETL
The former is postgresism, the latter is what SQL Server has (but…
- RETURNING clause to allow user to see what exactly was done and save on RTT
- WHEN NOT MATCHED BY [SOURCE|TARGET] to allow easy syncing with source - e.g., ETL
The former is postgresism, the latter is what SQL Server has (but…
Lads - check out our super serious (no) reel for Techcrunch Disrupt!
https://youtu.be/RxijVdlUBZk
Ps: no sound needed :)
https://youtu.be/RxijVdlUBZk
Ps: no sound needed :)
MySQL HeatWave Performance Comparison
- 11X faster than Redshift
- 9X faster than Snowflake
- 9X faster than BigQuery
- 3X faster than Synapse
https://www.oracle.com/mysql/heatwave/performance/
- 11X faster than Redshift
- 9X faster than Snowflake
- 9X faster than BigQuery
- 3X faster than Synapse
https://www.oracle.com/mysql/heatwave/performance/
Oracle
HeatWave MySQL Performance Comparisons
HeatWave MySQL on OCI outperforms top cloud databases.
A few PG hacks from Alibaba (thx to @olegkovalov)
1) PostgreSQL Graph Search Practices - 10 Billion-Scale Graph with Millisecond Response
https://www.alibabacloud.com/blog/postgresql-graph-search-practices---10-billion-scale-graph-with-millisecond-response_595039
2) Social Friend Relationship System Practice in PostgreSQL - Accelerating Queries of Positive and Negative Relationships
https://www.alibabacloud.com/blog/595043
1) PostgreSQL Graph Search Practices - 10 Billion-Scale Graph with Millisecond Response
https://www.alibabacloud.com/blog/postgresql-graph-search-practices---10-billion-scale-graph-with-millisecond-response_595039
2) Social Friend Relationship System Practice in PostgreSQL - Accelerating Queries of Positive and Negative Relationships
https://www.alibabacloud.com/blog/595043
Any of you in St. Petersburg now? Let me know, there is a surprise for you :)
https://postgresml.org/blog/postgresml-is-8x-faster-than-python-http-microservices/#why-latency-is-important
(thx to @olegkovalov)
(thx to @olegkovalov)
PostgresML
PostgresML is 8-40x faster than Python HTTP microservices – PostgresML
PostgresML's architecture gives it a huge performance advantage over traditional deployments when it comes to latency, throughput and memory utilization.
duckdb going mad...
UNION Type
UNION Type
CREATE TABLE messages(u UNION(num INT, error VARCHAR));FROM-first
INSERT INTO messages VALUES (42);
INSERT INTO messages VALUES ('oh my globs');
-- SELECT clause is optional, SELECT * is implied (if not included)BTW SQL standard already has special syntax for it:
FROM tbl;
-- first 5 rows of the table
FROM tbl LIMIT 5;
-- SELECT can be used after the FROM
FROM tbl SELECT l_orderkey;
-- insert all data from tbl1 into tbl2
INSERT INTO tbl2 FROM tbl1;
TABLE tbl;COLUMNS Expression
SELECT MIN(COLUMNS(*)), COUNT(*) from obs;List comprehension support
SELECT COLUMNS('val[0-9]+') from obs;
SELECT [x + 1 for x in [1, 2, 3]] AS l;
DuckDB
Announcing DuckDB 0.6.0
The DuckDB team is happy to announce the latest DuckDB version (0.6.0) has been released. This release of DuckDB is named “Oxyura” after the White-headed duck (Oxyura leucocephala) which is an endangered species native to Eurasia. To install the new version…