Original subtitles

Presenter: Let's talk a little bit about

the different types of storage

and connection modes in Power BI.

In this course, all of the data sources

that we'll connect to

will use import storage mode

but it's really important to understand

that Power BI actually supports several different types

of storage and connection modes

and that they're each useful

in different types of scenarios.

Import mode is the default storage mode

where tables are stored in memory within Power BI

and queries are fulfilled by cached data.

Import data sets are really useful

when you need fast query performance,

and the dataset will be less

than one gigabyte after compression.

These are also super handy when

the source data doesn't change frequently

and when you don't need restrictions on Power Query,

the data modeling, and then DAX functions.

Direct query tables are connected directly

to the source data and any queries,

meaning data requests, that are needed

are pushed back to the source

to be executed on demand.

This storage method is most commonly used

when data sets are too large

to be stored in memory

and the source data updates very frequently,

and reports must reflect the most recent data.

Additionally, this is commonly used

when a company policy states

that data can't be imported

but can only be accessed

from the original source.

Composite models, as you may

have already guessed by the name,

are models that contain a mixture

of both import and direct query modes

or are created from multiple direct query tables.

Composite models are useful in cases where you

want to really boost your data model performance

by setting appropriate storage types

for each table or combine a direct query model

with additional imported data,

or create a single model

from two or more direct query models.

And our last option here,

live connections are used to connect

to pre-published Power BI data sets

from sources like Power BI Service

and Azure Analysis services.

Live connections are great

because they create one data set that serves

as a central source of truth.

All team members can author

and create multiple different reports

from the same underlying source,

and it also helps multi developer teams work

more efficiently and seamlessly,

where one team member may build the model

and another team member works on the visualization.

As I mentioned at the start of this lecture,

we're gonna be importing all of our data sets

into Power BI's memory,

and we'll be using import mode.

From a little bit of a higher level,

part of your role as an analyst

or a Power BI developer

may be to recommend data connection strategies

based on the business case or the business need.

The point here is that

I don't wanna cover all of the intricacies

and nuances of each of these connection modes,

but rather, I really just wanna help

make you aware that there are different ways

that you can connect to data within Power BI,

and that the reasons

for using those are going to vary.

So each of these storage modes definitely

has different types of features and restrictions,

and limitations and upsides and considerations,

and it's really important to read more

about each type of these storage modes

prior to using it.

All right, so with that,

we're gonna jump into our next lecture.