Martin Berger (berx)
@martinberx.bsky.social
husband & father;
performance seeker;
Oracle ACE Alumnus;
proud member of @SYM_42;
Views are my own;
performance seeker;
Oracle ACE Alumnus;
proud member of @SYM_42;
Views are my own;
Reposted by Martin Berger (berx)
Not long to go! AskTOM Office Hours. Oracle experts will be onlinefor you. Next session starts 11 November at 19:00 UTC https://asktom.oracle.com/ords/r/tech/catalog/session-landing-page?p2_event_id=79923585559994835692303731425966476480
November 9, 2025 at 6:17 PM
Not long to go! AskTOM Office Hours. Oracle experts will be onlinefor you. Next session starts 11 November at 19:00 UTC https://asktom.oracle.com/ords/r/tech/catalog/session-landing-page?p2_event_id=79923585559994835692303731425966476480
Reposted by Martin Berger (berx)
From the vault. Those pesky dates as strings. https://connor-mcdonald.com/2016/05/13/those-pesky-dates-as-strings
November 5, 2025 at 6:08 PM
From the vault. Those pesky dates as strings. https://connor-mcdonald.com/2016/05/13/those-pesky-dates-as-strings
note to self:
for a proper overview about specific SQLs executions,
use DBMS_SQLDIAG.REPORT_SQL
let it write a zipfile to a DIRECTORY
use level => 'ALL'
enjoy additional content int the zip
(more than shown in
blogs.oracle.com/coretec/post...
)
for a proper overview about specific SQLs executions,
use DBMS_SQLDIAG.REPORT_SQL
let it write a zipfile to a DIRECTORY
use level => 'ALL'
enjoy additional content int the zip
(more than shown in
blogs.oracle.com/coretec/post...
)
blogs.oracle.com
November 5, 2025 at 10:31 AM
note to self:
for a proper overview about specific SQLs executions,
use DBMS_SQLDIAG.REPORT_SQL
let it write a zipfile to a DIRECTORY
use level => 'ALL'
enjoy additional content int the zip
(more than shown in
blogs.oracle.com/coretec/post...
)
for a proper overview about specific SQLs executions,
use DBMS_SQLDIAG.REPORT_SQL
let it write a zipfile to a DIRECTORY
use level => 'ALL'
enjoy additional content int the zip
(more than shown in
blogs.oracle.com/coretec/post...
)
Reposted by Martin Berger (berx)
To all my US friends, with regard to a parliament not convened.
In Austria, in 1933 the parliament became nonfunctional because all 3 presidents resigned, so for some time nobody had the authority to convene the parliament. This led to dramatic consequences.
In Austria, in 1933 the parliament became nonfunctional because all 3 presidents resigned, so for some time nobody had the authority to convene the parliament. This led to dramatic consequences.
November 1, 2025 at 12:38 PM
To all my US friends, with regard to a parliament not convened.
In Austria, in 1933 the parliament became nonfunctional because all 3 presidents resigned, so for some time nobody had the authority to convene the parliament. This led to dramatic consequences.
In Austria, in 1933 the parliament became nonfunctional because all 3 presidents resigned, so for some time nobody had the authority to convene the parliament. This led to dramatic consequences.
note to self: when investigating long PARSE time, don't only rely on SQL Monitor Reports (to check individual SQLs for parsing) - always crosscheck with ASH!
(SQL Monitor Report does not always show the full beginning of the SQL_IDs parsing)
(SQL Monitor Report does not always show the full beginning of the SQL_IDs parsing)
October 30, 2025 at 8:41 AM
note to self: when investigating long PARSE time, don't only rely on SQL Monitor Reports (to check individual SQLs for parsing) - always crosscheck with ASH!
(SQL Monitor Report does not always show the full beginning of the SQL_IDs parsing)
(SQL Monitor Report does not always show the full beginning of the SQL_IDs parsing)
next level: SQL_Optimizer trace?
October 27, 2025 at 8:02 PM
next level: SQL_Optimizer trace?
right now I have to deal with SQLs which I would not be able to store on a 5¼-inch floppy disk single sided ....
October 27, 2025 at 7:59 PM
right now I have to deal with SQLs which I would not be able to store on a 5¼-inch floppy disk single sided ....
Things I never thought I would ever read - today:
Logical SQL Reference Guide for Oracle Business Intelligence Enterprise Edition
docs.oracle.com/middleware/1...
¯\_(ツ)_/¯
Logical SQL Reference Guide for Oracle Business Intelligence Enterprise Edition
docs.oracle.com/middleware/1...
¯\_(ツ)_/¯
docs.oracle.com
October 22, 2025 at 8:11 AM
Things I never thought I would ever read - today:
Logical SQL Reference Guide for Oracle Business Intelligence Enterprise Edition
docs.oracle.com/middleware/1...
¯\_(ツ)_/¯
Logical SQL Reference Guide for Oracle Business Intelligence Enterprise Edition
docs.oracle.com/middleware/1...
¯\_(ツ)_/¯
Reposted by Martin Berger (berx)
Oracle 26ai¹ was released yesterday.
From the SQL standpoint it's worth mentioning that is has got the QUALIFY clause:
I bet this starts an avalanche (there are already PostgreSQL patches).
modern-sql.com/caniuse/qual...
¹ 26ai is the marketing version, 23.26.0 is the technical version.
From the SQL standpoint it's worth mentioning that is has got the QUALIFY clause:
I bet this starts an avalanche (there are already PostgreSQL patches).
modern-sql.com/caniuse/qual...
¹ 26ai is the marketing version, 23.26.0 is the technical version.
QUALIFY clause (window functions)
QUALIFY: filter rows based on window functions
modern-sql.com
October 15, 2025 at 3:36 PM
Oracle 26ai¹ was released yesterday.
From the SQL standpoint it's worth mentioning that is has got the QUALIFY clause:
I bet this starts an avalanche (there are already PostgreSQL patches).
modern-sql.com/caniuse/qual...
¹ 26ai is the marketing version, 23.26.0 is the technical version.
From the SQL standpoint it's worth mentioning that is has got the QUALIFY clause:
I bet this starts an avalanche (there are already PostgreSQL patches).
modern-sql.com/caniuse/qual...
¹ 26ai is the marketing version, 23.26.0 is the technical version.
Note to self:
the time you gain by faster typing you loose by searching for typos.
(no, they do not need to care about _JAVA_OPtIONS )
the time you gain by faster typing you loose by searching for typos.
(no, they do not need to care about _JAVA_OPtIONS )
October 2, 2025 at 1:21 PM
Note to self:
the time you gain by faster typing you loose by searching for typos.
(no, they do not need to care about _JAVA_OPtIONS )
the time you gain by faster typing you loose by searching for typos.
(no, they do not need to care about _JAVA_OPtIONS )
Reposted by Martin Berger (berx)
Our SIGMOD paper with our friends at Tsinghua + @wesmckinney.com + @pateljm.bsky.social on creating a next generation open-source data file format is out. F3 is a future-proof file format avoids the mistakes of Parquet.
📄 Paper: db.cs.cmu.edu/papers/2025/...
📁 Code: github.com/future-file-...
📄 Paper: db.cs.cmu.edu/papers/2025/...
📁 Code: github.com/future-file-...
October 1, 2025 at 1:49 PM
Our SIGMOD paper with our friends at Tsinghua + @wesmckinney.com + @pateljm.bsky.social on creating a next generation open-source data file format is out. F3 is a future-proof file format avoids the mistakes of Parquet.
📄 Paper: db.cs.cmu.edu/papers/2025/...
📁 Code: github.com/future-file-...
📄 Paper: db.cs.cmu.edu/papers/2025/...
📁 Code: github.com/future-file-...
A nice riddle:
what happened between these 2 screenshots?
workload is exactly the same,
eCPUs are constant,
instances are constant,
no change in data, tables, indexes, ...
hint: 23ai
what happened between these 2 screenshots?
workload is exactly the same,
eCPUs are constant,
instances are constant,
no change in data, tables, indexes, ...
hint: 23ai
October 1, 2025 at 8:34 AM
A nice riddle:
what happened between these 2 screenshots?
workload is exactly the same,
eCPUs are constant,
instances are constant,
no change in data, tables, indexes, ...
hint: 23ai
what happened between these 2 screenshots?
workload is exactly the same,
eCPUs are constant,
instances are constant,
no change in data, tables, indexes, ...
hint: 23ai
no, the last element in the filter
COLUMN IN ('bla', 'yadda', NULL)
will NOT do what you might had intended.
COLUMN IN ('bla', 'yadda', NULL)
will NOT do what you might had intended.
September 19, 2025 at 11:34 AM
no, the last element in the filter
COLUMN IN ('bla', 'yadda', NULL)
will NOT do what you might had intended.
COLUMN IN ('bla', 'yadda', NULL)
will NOT do what you might had intended.
Reposted by Martin Berger (berx)
Out of The Queue Comes ... Amdahl’s Law!
www.perfdynamics.com/Papers/amdsy...
www.perfdynamics.com/Papers/amdsy...
September 18, 2025 at 5:10 PM
Out of The Queue Comes ... Amdahl’s Law!
www.perfdynamics.com/Papers/amdsy...
www.perfdynamics.com/Papers/amdsy...
Reposted by Martin Berger (berx)
Previously on the blog. The performance report you are NOT using. https://connor-mcdonald.com/2021/04/30/the-performance-report-you-are-not-using/
September 17, 2025 at 6:08 PM
Previously on the blog. The performance report you are NOT using. https://connor-mcdonald.com/2021/04/30/the-performance-report-you-are-not-using/
Reposted by Martin Berger (berx)
In all the #Oracle23ai New-Feature-stories, how did this one got lost ??
The information in the new V$SQL_HISTORY is a Hidden Gem!
Can we give this thing a bit more attention?
Great Potential as a "Problem Fixer" for Devs, Ops, and DBA's...
simpleoracledba.blogspot.com/2025/09/orac...
The information in the new V$SQL_HISTORY is a Hidden Gem!
Can we give this thing a bit more attention?
Great Potential as a "Problem Fixer" for Devs, Ops, and DBA's...
simpleoracledba.blogspot.com/2025/09/orac...
Oracle23ai, and a new kind of history.
TL;DR: I discovered the view to find sql_history of a session. I can use that to report "what a session has done". Much more Simple than vi...
simpleoracledba.blogspot.com
September 17, 2025 at 3:34 PM
In all the #Oracle23ai New-Feature-stories, how did this one got lost ??
The information in the new V$SQL_HISTORY is a Hidden Gem!
Can we give this thing a bit more attention?
Great Potential as a "Problem Fixer" for Devs, Ops, and DBA's...
simpleoracledba.blogspot.com/2025/09/orac...
The information in the new V$SQL_HISTORY is a Hidden Gem!
Can we give this thing a bit more attention?
Great Potential as a "Problem Fixer" for Devs, Ops, and DBA's...
simpleoracledba.blogspot.com/2025/09/orac...
Little detail I read a long time ago (guess what book?)
and verified today:
"minimum active SCN", Delayed Block cleanout and ORA-1555.
oh what join in the little details, and how another "totally independent" session can kill your users experience.
and verified today:
"minimum active SCN", Delayed Block cleanout and ORA-1555.
oh what join in the little details, and how another "totally independent" session can kill your users experience.
September 11, 2025 at 11:04 AM
Little detail I read a long time ago (guess what book?)
and verified today:
"minimum active SCN", Delayed Block cleanout and ORA-1555.
oh what join in the little details, and how another "totally independent" session can kill your users experience.
and verified today:
"minimum active SCN", Delayed Block cleanout and ORA-1555.
oh what join in the little details, and how another "totally independent" session can kill your users experience.
Reposted by Martin Berger (berx)
🚀 New DEBUG=TRUE flag makes ORDS container debugging way easier! 🐳
No more hacks—see exactly what’s happening during startup in ORDS 25.2.3. 💡
Here’s how it works (with live example):
👉 martincarstenbach.com/2025/09/10/n...
#ORDS #DevOps #Oracle #containers
No more hacks—see exactly what’s happening during startup in ORDS 25.2.3. 💡
Here’s how it works (with live example):
👉 martincarstenbach.com/2025/09/10/n...
#ORDS #DevOps #Oracle #containers
New DEBUG flag simplifies troubleshooting ORDS container image issues
I previously wrote about troubleshooting the ORDS (Oracle REST Data Services) container’s entrypoint, e.g. the script that’s run whenever an ORDS container starts. This older article is…
martincarstenbach.com
September 10, 2025 at 3:02 PM
🚀 New DEBUG=TRUE flag makes ORDS container debugging way easier! 🐳
No more hacks—see exactly what’s happening during startup in ORDS 25.2.3. 💡
Here’s how it works (with live example):
👉 martincarstenbach.com/2025/09/10/n...
#ORDS #DevOps #Oracle #containers
No more hacks—see exactly what’s happening during startup in ORDS 25.2.3. 💡
Here’s how it works (with live example):
👉 martincarstenbach.com/2025/09/10/n...
#ORDS #DevOps #Oracle #containers
Reposted by Martin Berger (berx)
You can define your own inline JavaScript code delimiters in Oracle MLE — even emojis (💥).
A small but useful detail I explored after a great question from @oyvindisene.bsky.social at #POUG2025.
Blog post 👉 martincarstenbach.com/2025/09/08/i...
#oracle #MLE #javascript
A small but useful detail I explored after a great question from @oyvindisene.bsky.social at #POUG2025.
Blog post 👉 martincarstenbach.com/2025/09/08/i...
#oracle #MLE #javascript
Inline JavaScript and some further details about code delimiters
I recently attended POUG 2025, perhaps the most technical conference in Europe with focus on Oracle Database. My friend Øyvind Isene presented about Multilingual Engine (MLE)/JavaScript and made a …
martincarstenbach.com
September 8, 2025 at 9:19 AM
You can define your own inline JavaScript code delimiters in Oracle MLE — even emojis (💥).
A small but useful detail I explored after a great question from @oyvindisene.bsky.social at #POUG2025.
Blog post 👉 martincarstenbach.com/2025/09/08/i...
#oracle #MLE #javascript
A small but useful detail I explored after a great question from @oyvindisene.bsky.social at #POUG2025.
Blog post 👉 martincarstenbach.com/2025/09/08/i...
#oracle #MLE #javascript
Reposted by Martin Berger (berx)
Wow…can’t believe it took me 10 years to get to @pougorg.bsky.social but it delivered in every possible way. So great to see old friends and I haven’t learned that much at a conference in a very long time. ⭐️⭐️⭐️⭐️⭐️
September 7, 2025 at 1:22 PM
Wow…can’t believe it took me 10 years to get to @pougorg.bsky.social but it delivered in every possible way. So great to see old friends and I haven’t learned that much at a conference in a very long time. ⭐️⭐️⭐️⭐️⭐️
#POUG2025 is closing down.
Thank you @luizafrompoland.bsky.social , @ora600pl.bsky.social , all the sponsors, speakers, attendees and people in the background.
Hope to see you all at #POUG2026 !
Thank you @luizafrompoland.bsky.social , @ora600pl.bsky.social , all the sponsors, speakers, attendees and people in the background.
Hope to see you all at #POUG2026 !
September 6, 2025 at 2:40 PM
#POUG2025 is closing down.
Thank you @luizafrompoland.bsky.social , @ora600pl.bsky.social , all the sponsors, speakers, attendees and people in the background.
Hope to see you all at #POUG2026 !
Thank you @luizafrompoland.bsky.social , @ora600pl.bsky.social , all the sponsors, speakers, attendees and people in the background.
Hope to see you all at #POUG2026 !
Traditionally, the last presentations at #POUG are some easy stuff for an exhausted audience.
@christophlutz.bsky.social will for sure follow this tradition this year #POUG2025 😁
@christophlutz.bsky.social will for sure follow this tradition this year #POUG2025 😁
September 6, 2025 at 1:44 PM
Traditionally, the last presentations at #POUG are some easy stuff for an exhausted audience.
@christophlutz.bsky.social will for sure follow this tradition this year #POUG2025 😁
@christophlutz.bsky.social will for sure follow this tradition this year #POUG2025 😁
Upgrades are automagic nowadays - aren't they?
@dohdatabase.com tells us what (..) is going on under the hood - or what can be achieved manually.
@dohdatabase.com tells us what (..) is going on under the hood - or what can be achieved manually.
September 6, 2025 at 12:49 PM
Upgrades are automagic nowadays - aren't they?
@dohdatabase.com tells us what (..) is going on under the hood - or what can be achieved manually.
@dohdatabase.com tells us what (..) is going on under the hood - or what can be achieved manually.
@fsimondba.bsky.social will keep us awake with insights into MAA details.
Introduced by @ludovico.bsky.social at #POUG2025
Introduced by @ludovico.bsky.social at #POUG2025
September 6, 2025 at 11:48 AM
@fsimondba.bsky.social will keep us awake with insights into MAA details.
Introduced by @ludovico.bsky.social at #POUG2025
Introduced by @ludovico.bsky.social at #POUG2025
September 6, 2025 at 10:02 AM