Thanks Adam/Christian, this video is worth every second of the time spent

@GuyInACube

5 жыл бұрын

Love it! Thanks for watching!

Christian Wade is one of my favourite Microsoft speakers! Absolutely love his enthusiasm for the product.

@GuyInACube

5 жыл бұрын

I love working with Christian! He always brings the energy.

Christian is flawless in explaining this. Love the energy.

This is the video from 2018... How can it have only 600 likes?! Super cool, really glad I found it.

Great video, Adam and Christian! I can't wait to give this a go. Keep it coming...

@GuyInACube

5 жыл бұрын

Thanks Alex!

Great video, and explanation on how it all works. As always a few nuggets of gold from Christian

@GuyInACube

5 жыл бұрын

Thanks Gilbert! We had fun making this. A ton of good stuff inside it.

The video is a goldmine of information! Thanks as always :D

@GuyInACube

5 жыл бұрын

Agreed. I cringed at the 30 minute time, but there is sooooo much good stuff in there that I was ok with it.

Wow, thank you a lot! This feature is now so clear to me :)

you people deserve a 2 MILLION SUBS

Thanks for sharing. I guess I need another "Guy in a Cube" Video where Aggregations and the different Modes are used on a real life example? Is that possible? ;)

Great feature. Great Video. Thanks guys!

@GuyInACube

5 жыл бұрын

Thank you so much!

Hi Adam and Christian, Thanks for the video, super interesting! We usually work with SSAS cubes instead of Power BI datasets. Does something similar exist for SSAS-cubes?

Thank you for the video! so much helpful :)

@GuyInACube

4 жыл бұрын

Glad we could be of help to you. Thanks for watching! 👊

Thanks, great video love the possibilities.

@GuyInACube

4 жыл бұрын

Happy to help. Thanks for watching! 👊

Great feature in power bi. this will lift the power bi to the next level. Can you also let us know if the same feature also added to Azure Analysis Services in the near future?

Bang!!. Wonderful video guys. Thanks

@GuyInACube

4 жыл бұрын

Appreciate that! Thanks for watching 👊

Thanks a lot for this video Really useful, i have question, is it possible to create multiple aggregate tables?

It's good to see good old - and in many ways very useful - functionality from multidimensional models coming back to tabular model. I am looking forward to it! It was missed for a long time for the obvious reasons. About timelines ;-) soon will be this year? Is my assumption correct that it's coming at the same time in Azure Analysis Services?

@GuyInACube

5 жыл бұрын

for the soon piece, we announced that this is now available in the service! I'm not aware of any news on if this is coming to Azure AS.

Thanks for the great video! Is there a limitation when using DAX calculated tables (which by default are import/cached)? I have a very simple model with a Fact Table (DirectQuery), the Agg table (Import) and then the DateDim being a calculated table. The aggregation works until the moment I drag and drop the date from my DateDim. At that point it is still sending queries back to my SQL db. Since I can't convert a DAX table to Dual, is this a limitation or am I missing something?

Nice, Nice and Nice. I hope more videos for share quick, for various scenarios. and also advisor feature for aggregate, like recommend setting, like recommend improvement based existing report, and also information how reduce the CPU to DirectQuery table, Network for reduce cost of daily of DirectQuery table - aggregate seems to reduce CPU, Network and so on. Thank you Nice video.

@GuyInACube

5 жыл бұрын

Thank you! There was a ton of great info in this video. We will look at other scenarios related to this feature as we get more information.

Awesome!

christian wade couldnt stop laughing for YOooooo

Is this feature coming to SSAS Tabular in the future?

"I know Kung Fu". Whew, I'm going to watch this one multiple times to soak it in, but really cool possibilities. I like the term "balanced architecture" Christian used at the end. That may be the key phrase moving forward. Thank you for sharing.

@GuyInACube

5 жыл бұрын

Thanks for watching Lance! Agree on watching it a few times. So many good bits of info in this video!

Thanks for detailed description, I have a single table and visualization created in a report. Users want to export underlying data from this visualization!! How to do that I am not able to setup aggregate on it to accomplish pls help

can you please include a video on how to find max values across multiple columns the way we use in Excel - MAX(col1, col2, col3....)?

Hi guys. I'm running into an issue where I want to count distinct values in a column - but do not want to include the key (as it would hurt the aggregation as such). Do you have any pointers on how you would approach distinct count on aggs? thanks.

Is it possible to drillthrough from the aggregated level to the detailed level?

Great video. What's the minimum laptop configuration to run the data?

@GuyInACube

5 жыл бұрын

It depends. It depends on how much data you are working within the aggregate table. For the 1 trillion row demo, the PBIX was about 5GB compressed on disk and took about 22GB of memory for the Power BI Desktop process.

Wonderful! It was amazing presentation. Just one question I have. These aggregate tables are Duplicate Tables in PBI (Duplicate or Referenced)? Or they are directly fetched from source by firing aggregated query?

@marquesman87

4 жыл бұрын

Yes same doubt here. It's not really clear to me how the aggregated tables are built.

@project365

3 жыл бұрын

@@marquesman87 Yes, these aggs can be built at the data source or through M. This feature helps to redirect to the right table - detail or aggs table.

@AnyaAmasova1

2 жыл бұрын

at 5:23 they touch on this: 'Use Get Data and set up the relationships as usual'. (I personally just right-click and duplicate a query in Power Query, and use 'Choose Columns' to keep only the columns I want for the agg table).

Wonderful! Can I have multiple aggregation tables? Meaning: an agg table for a set of dimensions (product and store) and another for date? I'm doing a test but the more dimensions I add the bigger the agg table become, impacting the performance. Thanks :)

@AnyaAmasova1

2 жыл бұрын

Yes you can :) Check the video in this series 'Using Precedence to force Power BI Aggregations'. (I know this is an old comment, but maybe someone will watch this video today and ask themselves the same thing ;))

So how do you create Sales Agg table itself? Make a blank table and then configure it?

Mind blowing, absolutely mind blowing. I wish there was a SMASH button. I loved it. Thanks Christian and Adam.👌🏽 I have just one problem, well, I want this to be supported in Power BI Documenter. 🤔

@GuyInACube

5 жыл бұрын

Woot! Thanks Soheil! This was a longer video but so much great stuff inside of it.

Brilliant! ... I want ask: Is the Aggregation Table should be populated with aggregated dat in the database before pulling it in to Power BI model. If yes, can I can create the aggregation table as View?

@GuyInACube

5 жыл бұрын

That's an option. Different ways of doing it. From the SQL connector perspective, you can use a Table or View.

Thank you for sharing the knowledge! Clicky Clicky Draggy Droppy all the way.. let's crush this bloody Petabytes once for all

@GuyInACube

5 жыл бұрын

YES! Thanks for watching!

"Hopefully you found this useful" you suggest? YES ENORMOUSLY USEFUL! Thank you!

new desktop coming soon for power BI report server too???

Can you do a video to explain how to do cumulative or running total or sum of counts of text data. I have multiple years of data by years and months. The data is text, say “M”, “L” etc… in a column. I need to draw a chart showing both counts (bar) and running counts (line) by months. And from a slicer when I select a specific year then I should be able to visualize the counts and running counts of the text of the selected year. So running counts should be for a selected time say a year, not cumulative counts of all years starting from Min year to selected Max year.

Your Calendar table seems to have a calendar key, but appears to have a day granularity. Is there a way to take advantage of this if we want to aggregate by month rather than day?

@scottpowell5205

5 жыл бұрын

Hi Moof, it appears this is possible if you use a "snowflake" schema - for example, in this video they show aggregating at the product sub-category, which is a parent hierarchy level. But it's not clear to me if you HAVE to use a snowflake schema to accomplish that. The problem with snowflakes is you can't create hierarchies because the levels are in different tables.I'd love to know if there's a way to do this without snowflaking...maybe (???) an aggregate table on top of the date dimension? But there really isn't any documentation that's clear, and I'm not sure how to contact folks in the know.

Why the columns are used from main table even after created aggregated table and set precedence to higher , when run in dax studio it is showing no mapping and match result is failes so please advise where do i make chnages

I made a connection with zameen.com and extract property price on a regular basis. Now I have just 2500 rows but whenever I refresh, it usually takes 30 minutes which is very long. How it can be improved? Please guide.

Question: I have a ten year old fact table but i only want to aggregate the last 12 months cause in most of visuals i only show the data of last 12 months to the end-user. How can i do this?

Can we not just create a View on the Database these tables are sitting on which will do the same Aggregation by doing SUM and group by? What is the the difference between the two? View and Aggr

Info: Joining an imported agg table and a DQ dim table with key value does not work to match agg table. Only if both agg and dim table are DQ...

I'm curious - are aggregations some sort of "query plan" feature in Power BI - or was this an addition to the SSAS tabular model?

@venkateswarareddymarella4955

5 жыл бұрын

I believe this is part of performance improvements while using direct or import mode in desktop

@GuyInACube

5 жыл бұрын

It is a way to work with those larger datasets in a faster way. So, instead of having to get all the detail rows, even over something like DirectQuery - which could be slow for very large datasets, the aggregate table can be used to service the majority of visualization needs in a very fast way. Going against an aggregate will be way faster than over all of the detail rows. You tell Power BI the mappings between the detail table and the aggregate table and the engine takes care of the rest.

@scottpowell5205

5 жыл бұрын

Thanks Adam! One other question I've run in to as I start trying to play with this...I noticed that in addition to omitting dims completely in the aggregate table, he also chose to store the product dim at a parent, non-leaf level. To do this, he used snowflaked product dimension tables. Is there a way to do this without snowflaking? The big drawback with snowflaked tables is you can't build a Power BI hierarchy over them any more...so you lose drill functionality.I've got a use case right now where I'd like to store aggregations at the month level instead of day level...but I don't want to lose the ability to drill from year -> quarter -> month -> day.Thanks!Scott

@ScottSenkeresty

5 жыл бұрын

@@GuyInACube It really is brilliant. Just mad props to the team on this one.

this is kickass obviously, but quibble: you guys keep saying "there's no relationship!" - it looks to my eye like that Manage Aggregations screen DEFINES a relationship. A more abstract kind of relationship than simple dax, but a relationship nonetheless. that's like the POINT.

In the meaning of performance it's not good to connect tables in model by the key values. Better is to merge them in power query and set the aggregation directly to the dimProduct-ProductName (for example)