Video

watching june 2024....

Is that still accurate in 2024?

Blame China, correct? And NY Times reports that China corrupted elections in Canada. China also destroyed the free culture that had been Hong Kong.

Our savior

500ms foiled NK or China or Russia or CIA/Mossad plans

Who is here after XZ utils backdoor story?

great vid

Is this the same guy who found out the backdoor in xz?

*Promo sm*

if we could lower selectivity on the slides in this presentation, would be awesome! otherwise, great content as always.

Amazing content, so much I didint knew :D

The few The proud The marineß

USA 🇺🇸

😄

Citus is best

<3 I am so happy to see this become a reality. Have a multi tenant applicaiton, been running for year and looking at citus project with envy. Now I can finally make the switch

Promo SM

This was incredibly helpful to start digging into how RDBMS actually work. I thought the presenter picked his focus at each stage really well to show their relevance. I cant judge how outdated this is, but in 2023 it still seems to overlap with most of what I see in official PostgreSQL documentation. I encourage people to look at the docs themselves, because they go into much more detail after this nice starting point.

Insightful!

"Good code explains the problem, and solves it as a side effect" 💯 Great episode, thanks everyone ❤

Is this some scientology sh+t? Lol

Very interesting talk. Thank you for sharing.

𝓹𝓻𝓸𝓶𝓸𝓼𝓶

Am an avid user of this amazing tool, looking forward to using version 11 in a massive new project! Perfect timing guys.

In case of Sharding, how is primary key created? is it a clustered composite (tenant_id, campain_id) or a primary key on campain_id and index on (tenant_id, campain_id)? Thanks.

Great presentation, I hope I can pick some ideas from it on my next db design

Nice demo, thanks! I am going to try citus 11 for my current project connected with mobile analytics

hello

It feels so weird to have such a small citus data community , 64 views 0 comments T.T

Not only that the content is thorough, the speaker also did a phenomenal work expressing it in simple, digestible terms one at a time. Big shout out.

5:30 it's running Log4j tbh if your footprint can be small enough it sounds like a great idea to me to cohost your db and app on the same metal. What better way to not expose port 5432 by only using UNIX sockets?

If you don't define any distribution function that table will work as local table and data store into coordinator node. I can see coordinator node resource configuration less compare to worker nodes .Could you please share the results if coordinator and worker node both with same config.

good stuff but presenter is a bit anxious

DO NOT be rude. This dude knows his stuff & has delivered an excellent presentation. No need to get bitchy & personnel. Can you (Doco KD70C0) do better? Jerk.

postgres=# select * from pgbench_history; Time: 320.778 ms postgres=# SELECT create_distributed_table('pgbench_history', 'aid'); NOTICE: Copying data from local table... NOTICE: copying the data has completed DETAIL: The local data in the table is no longer visible, but is still on disk. HINT: To remove the local data, run: SELECT truncate_local_data_after_distributing_table($$public.pgbench_history$$) create_distributed_table -------------------------- (1 row) Time: 955.560 ms postgres=# select * from pgbench_history; Time: 1341.283 ms (00:01.341) ??? Time is increase! any other bench do the same

i have a question. what is the resource for single instance of postgres...?...in distributed we see 4 nodes with 8 vcpus and ~ 60 GB RAM each node? is single node 32 vCPus and 240 GB RAM or 1/4?

repo_id is not primary key - so how you have created it??

Why so expensive on Azure only? Why no on AWS, Heroku, DigitalOcean and so on?? It really sucks make solution monopolized by M$

Amazing that so few people asked questions. I have a few, in case anyone is reading these comments. 1. CPU wise, do you put any priority on L2 Cache size (the latest Epycs are way more expensive with 256 MB of L2 vs 128) ? 2. CPU cores vs Clock Speed. The frequency optimized Epycs are higher clock speeds (a little) but must more expensive. Worth it ? 3. RAM - Anything to say about CAS latency and clock speed ?

SO.

What about percentile aggregate, that can not be incrementally updated. can it ?

Excellent presentation. A very beautiful view of the 'internal workings' of PostgreSQL when you submit a query. A million lines of 'C' code. Isn't that amazing !! Like our human body. So many things need to come together and work in tandem to give us 'life'.

Thank you, I faced a problem while running a batch against postgres tables as I did big update on the whole table. VACUUM was the only way If I wanted to run two jobs concecutively...

Can you help me create database to manage property that use advanced postgreSQL?

I'm a bit confused because this article heap.io/blog/engineering/running-10-million-postgresql-indexes-in-production specifically says you have one event table per customer which sounds like you are not using the distributed table concept ?

Thank you for this introduction!

Worst English and monolithic voice really irritates

Is it possible can you share the scripts which you used . mail @ prazy.postgresdb@gmail.com

hello can we use raid 10 with ssd setup.

Still, nodes only make sense when you can't scale the SQL server any further. If you used the same RAM, CPU's etc on a single server it would outperform the distributed system. However, the price on smaller hardware may make it cheaper to build a distributed system as well, so keep this in mind when designing the system. The main point is this, NO SQL is not the solution based on scalability. No SQL databases are useful when the use cases supported the architecture of the DB in question. Different No SQL database, therefore, have different uses cases. If the use case is a SQL DB, this is, however, the way to go when scaling the single server further is not possible or too expensive. PS: Oracle RAC is doing a similar job, however for a premium license fee.