Видео

Good presentor, but I would prefer to read this.

Neither statefulsets or pods get ready 1/1

I never get the same results of minute 17:16

Pretty clear explanation !! Thankyou.

manyt rdbms like sqlserver columnar indesing also does the same thing

Hey! It would be great to see how it compares to in-memory data grids (GridGain/Ignite, Gemfire/Geode, Coherence) that provide some SQL-ish capabilities and are still commonly used for similar cases - muliti-QPS realtime analytics.

how to choose schema for table when creating table in the UI? after I create the schema in the UI then try to add realtime table there is no option so select the schema for the table.

Great Talk

This is the greatest intro song of all time.

Half of the stuff is wrong.

super cool visualization at the end for the star-tree index, really helped in understanding what that specific index is all about.

Is there no way to do table creation using a sql/hive-like statements/commands/scripts? My use case would be to use something similar to tables already created in hive...

I guess their comany StarTree is a Pinot service provider. Hence, to commoditize & sell their offering they would like to be baised towards Pinto😂 I mean nothing wrong in it from business perspective but open source fans will not love this. I don't think CliclHouse is that bad. Even in my company, we intentionally try to overpower serviceNow as we are serviceNow partners & service provider. Although there are other cheaper tools available in market to small use cases😂😂😂😂

It would be great if there is a cost calculator between those 4 options with few basic inputs like users, data volume, velocity etc. Another concern is ability to find resources.

When are you planning to release the grafana plugin ?

What about a spend insight dashboard?

Loving Flink so far and listening to Robert makes me realise why- it's so important to have a deep grasp of prior art and fundamentals. The insight that if you start with a streaming foundation you can implement batch over the top is so important. So often I see tools and systems try to add streaming later, and it just doesn't work. Whereas if you start pipelined, everything else is relatively easy. The hybrid source thing is so great, and I'm very happy never to have to write the plumbing for that myself ever again. Great interview, both of you.

I want to bring to your attention that your freelancers and distributors have been reported for using your advertising product to scam people worldwide. I urge you to take immediate action to ensure that your freelancers and/or distributors repay the money they have scammed from victims, and to promptly suspend the licenses of any scammers once you receive a report of their actions from the victims.

Amazing that this has no comments and only 70 ish views while there "data" and "AI" blasting from left and right 24/7

Hey, are you the head of a scamming group?

You’re so beautiful tho😊

I came across a combined task negative number and I can't find a solution what can be done

Hello, the combined task package consists of 2-3 common tasks. The commission for combined task packages is 3 times that of ordinary tasks. A combined task package consisting of 2 ordinary tasks means there will be one negative number, and a combined task package consisting of 3 ordinary tasks will have two negative numbers. This is your current last combined task, and you can cash it out on your own once completed.

They just don't know how clickhouse works

I'm a bit late to the Real-Time Analytics Podcast but I've enjoyed the episodes I've gotten to listen to over the last few months. Best of luck on your next adventure, Tim!

What sort of latency timing are we seeing from event ingestion at the start of the pipe to dashboard calculation update?

Thanks Tim sorry to see you go once again. Hope the hiatus is short but sweet. Looking forward to what’s next. Keep up the great work in advocating and educating us all. ❤

Redpanda is doing some really cool things. Agree with you Tim in that it blew my mind to hear the embedding use case by sending voice notes as messages and creating embedding as part of the WASM library. Great architecture pattern to have in the toolbox.

Did they get a reverse printed tree specifically for this video!?

Tim - The last 3 episodes (51, 52 and 53) were absolutely fantastic to understand the cutting edge of innovation real time data analytics ! Great job !

great

tim rocks 💥

baised

Awesome 🎉

What a fantastic show - such a pity that this has not gone viral yet. I'd love to have my kids listen to this actually, great history lessons...

to use Confluent and have transformation needs, the points mentioned are very interesting. Using Flink vs Kafka Stream, doing stateful VS stateless, are questions that we ask ourselves a lot in the company.

I just explained 5.25" and 3.5" floppies to a kid yesterday.

There aren't any proper video tutorials of Apache Pinot available, but if there are any, it would be a big help.

Thank you to all of the 70 Apache Pinot committers! Vector indexing is an awesome new feature.

Can't we simply utilize the API to update the table and schema?

code not working

1962 was also the first year of the Bell 103 modem. 300 bauds of POWAH!

Fun fact: the 1962 Tornados song "Telstar" was named after that first communications satellite.

Telstar!

sinking it*

Quine is so cool. Streaming graphs lend themselves to many use cases. From cybersecurity and networking to social media analytics.

Been loving this series so far such unspoken truths. Provide we can trust the source of truth. Great job had to comment. Can’t wait for the Pinot part.😊 it’s a journey through realtime. Love factual history

I was able to test this method via refering the pinot docummentation related to unnest. I believe if this is the same concept it misses certains steps. Though my question is does this feature apply to only JSON source data in kafka or can i apply this feature to other data formats like protubuf which is the format in my kafka topic.