Why u didn't use jupiter notebook in VSCode? It is more convenient than using terminal

is it the same as with incremental views that it depends on first table in join? and is only trigger for it? we needed to create a chain of mv's to get for whole view refreshed

Good that ClickHouse comes right out and says: ClickHouse is vehemently opposed to and condemns the war of aggression in Ukraine waged by Vladimir Putin and his regime.

A very interesting feature!

I have a question. If you are subscribed to a live data feed and writing into InfluxDB in real-time then you can create a real-time time series plot of the data in Grafana that will automatically update as new data arrives in the DB. Is this possible with Clickhouse and Superset? I’m specifically interested in real-time charts rather than static charts. Thanks.

For companies running multitenant infra or Saas companies isolatiing storage plane from compute this is a starting requirement, do you plan to make BYOC a standard feature after the launch ?

If I wanted to take all of the average prices by town and combine them to find the average price in the uk (where each down is weighted equally), is there a way to do that (presumably with a cascading materialized view)?

Congrats on the release! Great stuff. Cheers!

how far you are form a NATS Jetstream Tabel Engine?

Thanks! would be interesting to see a video how to streamline data from BigQuery to Clickhouse via clickpipes. There are no good examples so far, as far as I know

1

I'm not that familiar with American culture, and I thought Lauren was a girl name. I know that in Italy they use girl names for guys (like Andrea, Michele, Nicola) but wasn't aware this is something that Americans do as well.

that fking camera

Fantastic talk!

Nice video!

This guy sounds really smart! I like him

Audio is not great. But it is always insightful to listen to Alexy

Best video diving into CH’s shards and replicas 🎉

Would it be an idea , to run OPTIMIZE command post each upsert operation to ensure parts are merged?

Awesome content, but please record a audio in good quality or studio enviroment. it will help a lot for KZbin viewers.

Also a github repo would be nice because ClickHouse is underutilized in AI space. This solution would help people to adopt ClickHouse for AI use cases.

I'd appreciate to see another video showing the actualy presentation screen all the time, so we can also see what's happening on the screen. Thanks.

May you share slides please.

37:51 right at the very end, this is huge!! "You can even define Python code as user defined functions and reference them from the SQL language". Wow

How the parameters can be set to NULL? For example create table t(id int primary key, s String NULL); insert into t values(1, {PARAM1:String});

Not Coming Clickhouse option in dropdown

can you share more details about getRGB() and drawBlock() functions?

Interesting.. nice presentation

What’s the point of video if slides cant be focused

гае

Doby

Amazing talk! Complex but well spoken.

I can’t believe you delivered such a crisp talk while jet lagged after an 11 hour flight! ✈️ 😮

04:30 - Release 23.1 05:08 - SQL Language Features 05:16 - ARRAY JOIN with Map 07:15 - age function 09:27 - SQL Standard Hex And Binary Literal 11:47 - Extended Formats For Streaming 13:59 - Autodetect Headers of TSV/CSV 15:48 - Reading Structs from Parquet/ORC 17:11 - Reading from S3 21:30 - Data Parts Loading Optimization 23:09 - MMap for File Engine 26:07 - Parameterized Views 29:50 - Regexp Tree Dictionaries 34:45 - Query Result Cache 39:39 - Inverted Full Text Indices 43:07 - Integrations - Metabase 43:58 - Integrations - Java 44:23 - Integrations - Python 44:43 - ClickHouse Cloud 45:21 - Q&A

04:30 - Release 22.12 05:30 - SQL Language Features 06:00 - Numeric Literals With Undersore 07:10 - FROM able SELECT columns… 09:45 - GROUP BY ALL 11:38 - concatWithSeperator, concat_ws 14:55 - Decimal Operations With Specified Precision 15:56 - SQL UDF in CREATE Queries 21:02 - Operations & Monitoring 23:20 - system.moves table 23:50 - Prometheus Endpoint For ClickHouse Keeper 25:32 - Constraints For MergeTree Settings 26:50 - Security 27:00 - Password Complexity Rules 29:06 - Credentials Cleansing 30:36 - BSON Import/Export 33:30 - Grace Hash JOIN 39:30 - Compliance 41:45 - ClickHouse Cloud 43:40 - Q&A

06:30 - Release 22.11 07:00 - Composite Time Intervals 10:55 - Support For ** Glob 17:45 - Functions For Spark Compatibility 20:30 - Functions For Convenience 22:20 - Retries On INSERT 22:55 - Data Lakes 36:00 - S3 Readonly Disk 39:50 - Performance Improvements 42:00 - Bonus 44:44 - Integrations 45:55 - ClickHouse Cloud Beta 46:30 - QA

04:10 - Release 22.10 04:35 - Control Over Merging 07:15 - Resetting The Settings 09:33 - Fuzzy History Search 12:00 - Faster Reconnection to Keeper 14:10 - Keeper Snapshots in S3 16:05 - Backup To S3 17:25 - S3 WORM Disks 19:20 - Composable Protocols 28:00 - A Log For Asynchronous Inserts (`system.asynchronous_insert_log`) 29:22 - More Accept-Encodings for HTTP 30:22 - Morton Curve (`mortonEncode`, `mordonDecode`) 41:20 - Random Number Generation 44:45 - What Else? (ClickHouse Sink for Kafka Connect, clickhouse-kafka-connect) 46:05 - ClickHouse Cloud Beta 19:00 - QA

08:50 - Release 22.9 09:20 - UNION/INTERSECT/EXCEPT DISTINCT 10:40 - Insert Quorum 12:30 - Settings Changable For Readonly Users 19:15 - Saving Parsing Errors 22:40 - Asynchronous Initialization of Tables 25:40 - New And Improved JSON Formats 49:40 - How To Make ClickHouse Slow (Virtual Filesystems) 57:00 - Composable Cache 1:00:25 - Performance Improvements 1:01:50 - Experimental Feature: Vector Search (Annoy) 1:08:00 - KeeperMap Table Engine 1:08:15 - Integrations 1:09:35 - Bonus (custom dashboard) 1:13:00 - QA

Thank you for your very nice dbms ClickHouse

amazing...nice work..!

Some really amazing features here - as always. Thanks for leading the field of data storage and retrieval, thanks being the best.

do projections don't get used in replicatedMergeTree? let's say if i create a projection on table like select * order by <column_i_want>, <rest_of_primary_key_columns> will it be equivalent to adding that column in order by key in that table, if that didn't help, will creating a new table with my desired primary key help my queries, (i kind of chose suboptimal primary keys when creating the table but now we have ton of data and migration seem scary.), another stupid question, i am assume joins cannot be be used in projection statement. thanks.

How to read data in prometheus protocol from TimeSeries table?

Awesome!

Big Thumbs Up... Excellent short, precise and easy to understand video... I am looking forward to set these up with Click House Materialized View --> Endpoints... Thank You !

Great video! Very informative and to the point

if it helps, I'm really enjoying these videos

Hi. I followed the commands but cannot get "con.create_table( "flights", ibis.examples.nycflights13_flights.fetch().to_pyarrow(), overwrite=True )" to execute correctly. I get: "AttributeError Traceback (most recent call last) Cell In[2], line 3 1 con.create_table( 2 "flights", ----> 3 ibis.examples.nycflights13_flights.fetch().to_pyarrow(), 4 overwrite=True 5 ) File ~/.local/lib/python3.8/site-packages/ibis/examples/__init__.py:72, in __getattr__(name) 69 spec = _METADATA.get(name, {}) 71 if (key := spec.get("key")) is None: ---> 72 raise AttributeError(name) 74 description = spec.get("description") 76 _, ext = key.split(os.extsep, maxsplit=1) AttributeError: nycflights13_flights Running (version 24.6.1.2956 (official build)) locally. My locally running clickhouse executes faultlessly with all my other activities. Ubuntu 20.04. Python 3.8.10

Wow, kodak moment for all other olap stuff out there

Kql with ch is a game changer 🎉🎉🎉