Mauro Pagano
@mautro.com
I mess around with data.
Built some free Oracle tools at some point.
I like cats, a lot.
Built some free Oracle tools at some point.
I like cats, a lot.
Reposted by Mauro Pagano
"Oh, I'll just attend the webinar"
"Maybe I will download the slides"
Neither of that will EVER reproduce the value of sharing your knowledge and experiences face to face with your community colleagues at a meetup.
Support your local user group
"Maybe I will download the slides"
Neither of that will EVER reproduce the value of sharing your knowledge and experiences face to face with your community colleagues at a meetup.
Support your local user group
November 26, 2025 at 7:32 AM
"Oh, I'll just attend the webinar"
"Maybe I will download the slides"
Neither of that will EVER reproduce the value of sharing your knowledge and experiences face to face with your community colleagues at a meetup.
Support your local user group
"Maybe I will download the slides"
Neither of that will EVER reproduce the value of sharing your knowledge and experiences face to face with your community colleagues at a meetup.
Support your local user group
Reposted by Mauro Pagano
I just extended the Black Friday sale to the whole month of November!
All my existing full-week flagship courses are heavily discounted for the entire month, and more content coming soon!
tanelpoder.com/posts/black-...
All my existing full-week flagship courses are heavily discounted for the entire month, and more content coming soon!
tanelpoder.com/posts/black-...
November 2, 2025 at 9:45 PM
I just extended the Black Friday sale to the whole month of November!
All my existing full-week flagship courses are heavily discounted for the entire month, and more content coming soon!
tanelpoder.com/posts/black-...
All my existing full-week flagship courses are heavily discounted for the entire month, and more content coming soon!
tanelpoder.com/posts/black-...
Reposted by Mauro Pagano
The first deep dive in my new special-focus training series starts on Wednesday:
"Handling the Optimizer, Statistics and Workloads on Partitioned Tables in Oracle Databases" (5-6 Nov 2025 + Q&A followup sessions).
One demo table has 100k partitions ;-)
Sign up here:
tanelpoder.com/seminar/
"Handling the Optimizer, Statistics and Workloads on Partitioned Tables in Oracle Databases" (5-6 Nov 2025 + Q&A followup sessions).
One demo table has 100k partitions ;-)
Sign up here:
tanelpoder.com/seminar/
October 31, 2025 at 4:44 AM
The first deep dive in my new special-focus training series starts on Wednesday:
"Handling the Optimizer, Statistics and Workloads on Partitioned Tables in Oracle Databases" (5-6 Nov 2025 + Q&A followup sessions).
One demo table has 100k partitions ;-)
Sign up here:
tanelpoder.com/seminar/
"Handling the Optimizer, Statistics and Workloads on Partitioned Tables in Oracle Databases" (5-6 Nov 2025 + Q&A followup sessions).
One demo table has 100k partitions ;-)
Sign up here:
tanelpoder.com/seminar/
Reposted by Mauro Pagano
We published a new deep dive by Laurens Kuiper, who recently redesigned DuckDB's sort.
One data point: ordering the TPC-H SF100 lineitem table with the memory limit set to 30 GB is 3× faster in DuckDB v1.4 than in v1.3.
Read more at duckdb.org/2025/09/24/s...
One data point: ordering the TPC-H SF100 lineitem table with the memory limit set to 30 GB is 3× faster in DuckDB v1.4 than in v1.3.
Read more at duckdb.org/2025/09/24/s...
Redesigning DuckDB's Sort, Again
After four years, we've decided to redesign DuckDB's sort implementation, again. In this post, we present and evaluate the new design.
duckdb.org
September 25, 2025 at 6:35 PM
We published a new deep dive by Laurens Kuiper, who recently redesigned DuckDB's sort.
One data point: ordering the TPC-H SF100 lineitem table with the memory limit set to 30 GB is 3× faster in DuckDB v1.4 than in v1.3.
Read more at duckdb.org/2025/09/24/s...
One data point: ordering the TPC-H SF100 lineitem table with the memory limit set to 30 GB is 3× faster in DuckDB v1.4 than in v1.3.
Read more at duckdb.org/2025/09/24/s...
Reposted by Mauro Pagano
I have published the CatVector embedding-vector heatmap visualization app (for learning and fun!) as a static webpage now (just browser is enough, no downloads or Python needed).
And now we have dark mode too! 😼
tanelpoder.com/catvector/
And now we have dark mode too! 😼
tanelpoder.com/catvector/
August 25, 2025 at 1:32 AM
I have published the CatVector embedding-vector heatmap visualization app (for learning and fun!) as a static webpage now (just browser is enough, no downloads or Python needed).
And now we have dark mode too! 😼
tanelpoder.com/catvector/
And now we have dark mode too! 😼
tanelpoder.com/catvector/
Reposted by Mauro Pagano
As I'm creating new content for my latest online training courses, I just bundled the 2021 "Troubleshooting Very Complex Oracle Performance Problems" videos into the AOT & SQL Tuning courses, without an extra fee; existing attendees will have access too!
learn.tanelpoder.com
learn.tanelpoder.com
August 17, 2025 at 9:54 AM
As I'm creating new content for my latest online training courses, I just bundled the 2021 "Troubleshooting Very Complex Oracle Performance Problems" videos into the AOT & SQL Tuning courses, without an extra fee; existing attendees will have access too!
learn.tanelpoder.com
learn.tanelpoder.com
Reposted by Mauro Pagano
Reposted by Mauro Pagano
Optimizing eBPF I/O latency accounting when running 37M IOPS, on 384 CPUs
tanelpoder.com/posts/optimi...
tanelpoder.com/posts/optimi...
April 30, 2025 at 7:08 AM
Optimizing eBPF I/O latency accounting when running 37M IOPS, on 384 CPUs
tanelpoder.com/posts/optimi...
tanelpoder.com/posts/optimi...
Reposted by Mauro Pagano
0x.tools xCapture v3: Linux Performance Analysis with Modern eBPF and DuckDB 👀👀
tanelpoder.com/posts/xcaptu...
tanelpoder.com/posts/xcaptu...
April 23, 2025 at 11:15 PM
0x.tools xCapture v3: Linux Performance Analysis with Modern eBPF and DuckDB 👀👀
tanelpoder.com/posts/xcaptu...
tanelpoder.com/posts/xcaptu...
20y and 2 days ago (forgot!) I joined Oracle Italy…
April 8, 2025 at 5:09 AM
20y and 2 days ago (forgot!) I joined Oracle Italy…
Reposted by Mauro Pagano
We teamed up with the Disseminate: The Computer Science Research Podcast to launch a new series called DuckDB in Research.
In the first episode, show host Jack Waudby talks to Till Döhmen about DuckDQ, a Python library for data quality checks in ML pipelines: open.spotify.com/episode/2a0b...
In the first episode, show host Jack Waudby talks to Till Döhmen about DuckDQ, a Python library for data quality checks in ML pipelines: open.spotify.com/episode/2a0b...
March 15, 2025 at 8:36 PM
We teamed up with the Disseminate: The Computer Science Research Podcast to launch a new series called DuckDB in Research.
In the first episode, show host Jack Waudby talks to Till Döhmen about DuckDQ, a Python library for data quality checks in ML pipelines: open.spotify.com/episode/2a0b...
In the first episode, show host Jack Waudby talks to Till Döhmen about DuckDQ, a Python library for data quality checks in ML pipelines: open.spotify.com/episode/2a0b...
Reposted by Mauro Pagano
The total time to perform a cold run of the TPC-H queries on SF300 shows an almost 3× speedup between DuckDB 1.0.0 and DuckDB 1.2.0. The benchmark was run on a MacBook Pro (M1 Max / 64 GB RAM).
March 9, 2025 at 7:24 AM
The total time to perform a cold run of the TPC-H queries on SF300 shows an almost 3× speedup between DuckDB 1.0.0 and DuckDB 1.2.0. The benchmark was run on a MacBook Pro (M1 Max / 64 GB RAM).
Reposted by Mauro Pagano
Starting with version 1.2.0, DuckDB supports reading and writing Parquet Bloom filters. For details, read the blog post by @hannes.muehleisen.org: duckdb.org/2025/03/07/p...
March 7, 2025 at 11:36 AM
Starting with version 1.2.0, DuckDB supports reading and writing Parquet Bloom filters. For details, read the blog post by @hannes.muehleisen.org: duckdb.org/2025/03/07/p...
Reposted by Mauro Pagano
DuckDB 1.2.1 is out: github.com/duckdb/duckd...
March 5, 2025 at 9:45 PM
DuckDB 1.2.1 is out: github.com/duckdb/duckd...
Reposted by Mauro Pagano
Ready to play with Apache Arrow using the @duckdb.org Airport community extension? Check out #quackflight for a sample Flight server implementation supporting parallel inserts, ATTACH and various ticket examples interacting with #DuckDB github.com/quackscience...
#Apache #Arrow #Flight #DuckDB #OLAP
#Apache #Arrow #Flight #DuckDB #OLAP
GitHub - quackscience/quackflight: DuckDB API Server with Arrow Flight SQL Airport support and concurrent writes/reads (quackpipe)
DuckDB API Server with Arrow Flight SQL Airport support and concurrent writes/reads (quackpipe) - quackscience/quackflight
github.com
February 28, 2025 at 11:45 PM
Ready to play with Apache Arrow using the @duckdb.org Airport community extension? Check out #quackflight for a sample Flight server implementation supporting parallel inserts, ATTACH and various ticket examples interacting with #DuckDB github.com/quackscience...
#Apache #Arrow #Flight #DuckDB #OLAP
#Apache #Arrow #Flight #DuckDB #OLAP
Are people here also still on Twitter? If yes, why? Really, why?
January 22, 2025 at 6:00 PM
Are people here also still on Twitter? If yes, why? Really, why?
Reposted by Mauro Pagano
Folks: Stop making fun of JavaScript.
Meanwhile, JavaScript:
Meanwhile, JavaScript:
January 22, 2025 at 9:16 AM
Folks: Stop making fun of JavaScript.
Meanwhile, JavaScript:
Meanwhile, JavaScript:
Reposted by Mauro Pagano
oh, youre here from tik tok? (chuckle's lightly ) Is that like reddit for babies
January 19, 2025 at 3:27 PM
oh, youre here from tik tok? (chuckle's lightly ) Is that like reddit for babies
Name one thing that will never come to Oracle 😞
2025 is shaping up to be a breakout year for fast query result transfer with Apache Arrow. But what exactly makes it so fast? David Li, Matt Topol, and I break it down in this new blog post: arrow.apache.org/blog/2025/01...
How the Apache Arrow Format Accelerates Query Result Transfer
Arrow speeds up query result transfer by slashing (de)serialization overheads. We outline five key attributes of the Arrow format that enable this.
arrow.apache.org
January 14, 2025 at 4:57 AM
Name one thing that will never come to Oracle 😞
Reposted by Mauro Pagano
Does anybody know why DBMS_CLOUD installed on-prem can’t dump PQ files? Beside the obvious “no quick way to take data out of Oracle” mantra
December 16, 2024 at 4:40 PM
Does anybody know why DBMS_CLOUD installed on-prem can’t dump PQ files? Beside the obvious “no quick way to take data out of Oracle” mantra
Reposted by Mauro Pagano
Don’t you DARE disrespect the speed of @duckdb.org
December 13, 2024 at 3:50 PM
Don’t you DARE disrespect the speed of @duckdb.org
Reposted by Mauro Pagano
A sneak peek: the DuckDB CLI client's new nightly build now pretty-prints large numbers by default.
December 2, 2024 at 8:51 PM
A sneak peek: the DuckDB CLI client's new nightly build now pretty-prints large numbers by default.
Reposted by Mauro Pagano
Reposted by Mauro Pagano
In other, non-DuckDB news...
Snowflake continues to be the most expensive way to shuffle bytes from one S3 bucket to another S3 bucket.
Snowflake continues to be the most expensive way to shuffle bytes from one S3 bucket to another S3 bucket.
November 29, 2024 at 10:27 PM
In other, non-DuckDB news...
Snowflake continues to be the most expensive way to shuffle bytes from one S3 bucket to another S3 bucket.
Snowflake continues to be the most expensive way to shuffle bytes from one S3 bucket to another S3 bucket.