Welcome to SkillTech’s official Microsoft Azure DP-900 course series!
In this video, we dive deep into non-relational data—what it is, how it's different from relational data, and how Azure supports it through services like Azure Cosmos DB, Blob Storage, and more.
🔍 What you’ll learn in this session:
Introduction to non-relational data types (document, key-value, graph, columnar)
Core concepts and use cases of non-relational data
How Azure handles non-relational data with real-world examples
Overview of Azure Cosmos DB and other NoSQL storage options
Whether you're just starting your cloud journey or preparing for the DP-900: Microsoft Azure Data Fundamentals certification, this module will strengthen your foundational understanding of modern data structures.
Explore our other Courses and Additional Resources on: https://skilltech.club/
In this video, we dive deep into non-relational data—what it is, how it's different from relational data, and how Azure supports it through services like Azure Cosmos DB, Blob Storage, and more.
🔍 What you’ll learn in this session:
Introduction to non-relational data types (document, key-value, graph, columnar)
Core concepts and use cases of non-relational data
How Azure handles non-relational data with real-world examples
Overview of Azure Cosmos DB and other NoSQL storage options
Whether you're just starting your cloud journey or preparing for the DP-900: Microsoft Azure Data Fundamentals certification, this module will strengthen your foundational understanding of modern data structures.
Explore our other Courses and Additional Resources on: https://skilltech.club/
Category
🤖
TechTranscript
00:00okay now it's time to see our module number three where we are going to
00:13explore non-relational data in a short obviously non-relational data is going
00:18to be semi-structured or unstructured data and in this also we have something
00:23like a sure storage tables and all will focus on that the agenda of this module
00:28is having three lessons actually the first lesson is going to focus on
00:33exploring non-relational data services in a short so what kind of service options
00:38are available in that will focus on that lesson two is going to focus on how we
00:42can provision and deploy this kind of non-relational data services in a short
00:47so like in the previous module we have provision sequel server database in this
00:52we'll see something like a sure storage account cosmos DB kind of databases which
00:57will be non-relational data services in a short and then in the third and the
01:01final lesson of this module we will focus on how we can manage non-relational data
01:06stores in a short in this lesson we will explore non-relational data services in
01:12a short and when we talk about non-relational data services ensure there
01:16are two main services which are coming in that as your storage account and as your
01:21cosmos DB the objective of this particular lesson is first we need to figure out
01:26as your stable storage as your blob storage and ensure file storage which
01:31are actually coming under as your storage account and we will see that what
01:36kind of use cases will be there when we are going to use this kind of data formats
01:42and then we will also see what kind of benefits we are getting in that after this
01:46we will focus on ensure cosmos DB account which is Microsoft's own multi-model
01:51database this is not a full RDBMS kind of database because we are keeping this
01:57thing in non-relational category and what kind of new changes are there what
02:01kind of benefits are there associated with ensure cosmos DB that also we are
02:06going to focus in this lecture now let's start with exploring ensure table
02:11storage as I see this is one of the part of a sure storage account and ensure
02:16table storage allows you to store data in a tabular format now tabular format
02:21means you will have rows and column kind of structure in this and all the rows in
02:26the table must have a key and this is not something like a primary key this is
02:32going to be a unique key which is associated with this assure table storage
02:36and with every key you are going to have a value which is nothing but your data
02:40which you want to store with that as you can see on screen I have a customer ID
02:45which is my unique key associated with this assure table and then we have a
02:50values which can be in any kind of a structural format we can have different
02:55kind of comma separated values in that key it can be a string kind of key which
02:59we can put into that the difference between this table and a relational database
03:05table actually is this the Azure table storage has no concept of relationship so
03:10there is no relationship between one table and another table we also do not have
03:15concept of stored procedures or index six kind of thing into that also because I'm
03:19saying there is no relationship there will be nothing like foreign key in this
03:23inside of foreign keys primary keys and all we have key value pair kind of
03:28structure which we need to follow and then we have some different kind of keys
03:32which are known as partition key and raw key which will make sure that there's your
03:37table data is going to be properly organized we'll see this thing when we are
03:42going to see the demo for this lecture after this we have a sure blob storage
03:46now obviously I'm sure you heard this word blob earlier because blob stand for
03:51binary large object and this allows you to store data in the binary format
03:56directly obviously for computers and for clouds blob is a preferable format
04:02whenever you have to store huge data and you have to retrieve the data you want to
04:08transfer data from one place to another place blob is a preferable format for
04:12that and that's one of the reason I think almost all the clouds are going to
04:17support blob in Azure blob storage we have three types of blobs which are
04:22available we have block blob page blob and append blobs block blobs are going to
04:29have a maximum size of 4.7 terabyte and then this is actually perfect when you want
04:35to store huge amount of data in the binary format and then you can also access it
04:39infrequently as you want most of the time in the block blobs we are going to
04:45store data like images videos or some other bigger files which I want to store
04:50there as a binary block object and then I can just retrieve it as when I want
04:57same like that we have page blobs which are also some kind of a block block
05:03kind of thing but page blobs are actually holding up up to 8 terabyte of data
05:07maximum it is organized as something like a pages and that's why it's like it's
05:13going to have a collection of fixed size 512 bytes of pages into that you can
05:19think about a virtual disk or virtual hard drive kind of thing which you have
05:23used in your virtual machines page blobs are exactly like that where you can
05:27store data in the directly file format which will be treated with blobs but this is
05:32something which is going to be counted like page blobs and then we have a third
05:36and another interesting blob which is append blob which is perfect for
05:40scenarios where you have some log files some CSV or text files which you are
05:46going to store in this you want to make sure that the data size is huge but you
05:51also want to make sure that when you're storing this kind of huge data then this
05:55is just going to append a new data into the existing file like when you're
06:00storing logs you're not going to update the existing logs you just keep adding
06:04new new logs into that and that's what append blob is actually all about this
06:09maximum size for this one is going to be 195 GB which is I guess enough for any
06:16particular append files kind of structure after table and blob now we
06:22have a third one which is also part of Azure storage account which is known as
06:25as your file storage the file storage is actually going to create file shares name
06:31itself is adjusting that this allows me to upload files and then I can share it
06:36with my multiple cloud apps or multiple machines which are located anywhere this
06:42creates a file share in the cloud and it provides the ability to access the file
06:46share from anywhere with a internet connection in short this is going to
06:50transfer the files from here to any other place using internet the transfer it
06:56is really high it can share up to 100 terabytes of data in a single storage
07:01account and it is also secure because it is using SMB 3.0 protocol which maybe you
07:09know widely used by nowadays most of the file sharing mechanisms like Dropbox
07:14OneDrive Google Drive these all are using SMB 3.0 protocol now this is going to be a
07:20fully managed service data is going to be replicated locally and it is also going
07:24to be encrypted at rest when you want to share the files from one place to
07:30multiple places or multiple applications or multiple machines maybe it's
07:35located anywhere in this particular geographical location then also you can
07:40access the files using the assure file storage now after this the final one
07:45which is Azure Cosmos DB Azure Cosmos DB is a multi-model those equal database
07:51management system now why the world multi-model is associated with this
07:55database I will tell you that thing that when you're dealing with Azure Cosmos DB
08:00you can use various API's which are provided by Microsoft you can use a sure
08:07table API you can use a sequel API you can also use MongoDB or Cassandra API when
08:15you're dealing with Cosmos DB you want to treat it like a simple database which is
08:20just having key value pairs stored in the JSON files kind of thing you can use it
08:25like that or you want to treat it like a graph DB then you can use it with the
08:30gremlin API and it will be behave like a graph DB also and that's the reason this is a
08:34multi-model database it manage the data as a partition set of documents and most of
08:40the time the documents are going to be in JSON format this is allowing you to do
08:45real-time access with the really fast read and write latencies the throughput
08:50which is associated with Cosmos DB is going to give you millions of requests
08:55per second and I'm serious about this the minimum throughput in Cosmos DB
08:59starting from 400 requests per second the maximum can go up to 1 million
09:05requests per second per database inside one Cosmos DB account and this is a
09:11really fast performance it takes an advantage of Azure scaling and storage
09:16capabilities and it is also allowing you to do multiple regional replication with
09:21multi-legion write facility available for that so unlike a traditional SQL server
09:27on Azure where you can only have read only copies available geographically here
09:31you can have multi-region write which works with the highest throughput in the
09:36market if you ask me in which scenarios Cosmos DBs are going to be a perfect
09:41choice well there are a couple of actually if I show you some of them then
09:45the first one which I want to add on this if you have any websites which is
09:49e-commerce website or retail website where you are actually dealing with the
09:53products and millions of users accessing that thing you can use a sure Cosmos DB
09:58multi master application model along with Microsoft performance commitment
10:03data engineers can implement a data architecture to support web and mobile
10:08applications both that achieve less than 10 millisecond response time anywhere in
10:14the world and which is very good you can also use this thing for gaming and
10:19storing the data of the gaming applications again maybe a web or a mobile
10:23application modern games performs graphical processing in short the GPU has
10:29to be associated with that on mobile and console client and but they rely on the
10:34cloud to deliver customized and personalized content like in-game status
10:38social media integrations or even the high-score leaderboards for all this data
10:42you can use Cosmos DB and the third but the important one is IOT scenarios if you
10:49have any internet of things kind of devices are which are connected with one
10:54centralized IOT hub and then you want to actually capture the real-time data which
10:59is associated with that and you want to provide some kind of analytics you want
11:04to do for the processing on that then this sensor generated data can be also
11:09utilized and stored inside Cosmos DB so there are a couple of places where you can
11:14use this thing there are a couple of social media websites where we can apply Cosmos DB
11:19multiple scenarios are there you have to keep in mind when you have a higher
11:23throughput requirement and you want to still keep it in a much organized way then
11:30you can go with Azure Cosmos DB so in this lecture we have seen what kind of
11:34non-relational data services are available in Azure and we have mainly focus on
11:39Azure storage account and Azure Cosmos DB in storage account we have Azure
11:44tables we have Azure files and we have Azure blob kind of storage while Cosmos DB is
11:50a multi-model database which can allow me to store data in a JSON files with the
11:55key value pair format