Walkover to Azure Cloud Storage
11K views
Nov 7, 2023
Walkover to Azure Blob/ Tiered Storage, Fundamentals Of Azure Storage.
View Video Transcript
0:00
So Deepak, it's your turn now. Thank you everyone for listening to my presentation
0:08
This session will be available in Channel 9 this next week. Thank you
0:15
Thanks Rahat. So thanks everybody for having this our Tech Discussion webinar
0:27
webinar, I will try to interact, make it interactive as much as possible as well as my sole objective
0:36
is today after this session, we have fair idea about cloud storage, about blob storage
0:47
about data warehouse which we talked earlier and we came to know with the answers why we
0:53
need blob storage or why not we use another storage type. So that's my agenda
1:01
and thanks again. Today we talk about Azure Storage. As we know storage is the
1:10
most important part for any business right. If we have any business without
1:17
data or we can't afford to have data compromised in any condition. We can't
1:25
afford to have data still. We can't afford to have data having some network
1:32
compliances which is vulnerable to various regions. So we will cover all
1:40
that scenarios. So today we will discuss about how we can deliver the high scale
1:48
and low cost Azure storage data. We will do a live demo over here. I will share
1:55
sample code with you guys from which you can play around. That is
2:02
Visual Studio Code. You can download from my article from my website. We will look
2:09
our Azure storage tool also how we can have blob storage or unstructured data
2:19
how we can manage that how you know we can move ahead with that so that is we
2:28
are going to cover so first of all which I what I want to say like last time we
2:36
talk about the Azure Data Warehouse which is the structured way of storage
2:42
If I'm saying structured way you could say okay relational database maybe SQL
2:48
server. But today we will talk about blob storage which is another form of
2:55
storage which is unstructured database. If we talk about unstructured database probably things come into our mind what does that mean? That means if we have
3:10
data from satellite and we are storing it say live traffic data in Vancouver or
3:17
in Toronto or in Regina so that storage type would be unstoraged type. Another
3:26
Another storage type is having the log files from nuclear reactor where we are
3:35
updating it, appending it with infinite time, with infinite length. So that
3:44
would be the blob storage. Today we are not talking about structured storage which we did earlier but it would be the blob storage or
3:54
unstructured stories like log file, like image, like light telecast, recording etc
4:03
We will cover couple of scenarios to understand it better. We will cover the
4:09
key concept, what architecture it follows and we will going to have a demo, live
4:18
demo through Azure portal A, B by code, C by a tool. So the whole understanding
4:28
after this scenario after this session definitely we would have you know cover
4:34
about Azure storage in nitty-gritty and that that would be the objective of this
4:41
session. Okay so if I talk about current storage challenge, say being a
4:49
developer when I was developing an application 12 years back the main
4:55
challenge in front of me was to create a website having a database but I don't
5:04
know about what the scalability database required. After two years might currently I develop a web application having user base in Canada but after
5:17
three years if all the user base is shifted to the Europe then what will be
5:22
happen scalability being doing scalable in database that is really cumbersome task for any any developer or any team and that is also covered some risk that
5:35
cost us because we have to increase the infrastructure and without any
5:40
scalability limit business can't move forward. So if I would say all these
5:46
things are handled automatically by Azure Cloud Storage I think that would be a
5:54
great relief for any developer, any architect and from business perspective that is you know kind of business need. So that we are going to do with storage
6:09
and it adds the storage scalability limit. We don't have to worry about hard
6:16
drive, it is the automatically append. We don't have to worry about the risk
6:22
management nowadays what the threat we experience so that is also covered by
6:29
risk management and as far as we talk about the cost it is paper use so if you
6:35
are not using one PB of data or you are only using 100 GB of data then you will
6:43
be charged for 100 GB so that is the strength. So how the challenge can be
6:51
addressed by the storage challenge? We have limitless scalability, we have trust
6:56
because till now till today May 27th not a single incident has been reported of any threat compromised at cloud storage cloud economics and that gives to the business agility If we have limitless scalability we know if
7:17
today my application is at Canada or North America and tomorrow it's going to
7:22
be after a year in Europe or after that in Africa then definitely I have that
7:28
strength to expand my business without taking care of about the scalability. It
7:33
has trust it's paper use if I'm using 10 GB of data I will charge for 10 GB
7:41
not for 100 GB and that gives us the business agility that strength our
7:45
business so what is the Azure storage if we talk about Azure storage Azure
7:52
storage is the foundational building block of Azure like we talked about last
7:57
time we talked about Azure Data Warehouse where we store SQL, we have
8:04
data lake store but today we talk about Xbox, Skype, unstructured data. It is very
8:13
hyperscale, more than 120 trillion object. It's durable, it's secure, it's highly
8:20
available and it's open. We have APIs from which we can access the data, we can
8:26
play around that we can append it we can put it we can delete it we are going to
8:31
address and touch base those on further slides and we have strength of hybrid
8:39
stack also so that would be the azure storage azure storage services come into
8:45
two forms if we now we have fair about idea about azure storage where we talk
8:53
about that data storage or structured storage versus unstructured storage which is today we are covering so it's come into two form infrastructure as a service
9:04
where we have storage we have VM we have network or we have platform as a
9:09
service where we have existing frameworks say our web application and we
9:14
are using their service or compute power to have that so for Azure storage
9:21
service we have disk, we have files, couple of data which we need. Where a
9:27
platform as a service we have objects, tables and queues. So whatever our business
9:33
needs is, whatever the budget is as per our business scenario. If we are
9:37
appending our old application or versus if we are creating the architect of new
9:43
application we have all the strengths as IIS infrastructure as a service and
9:50
platform as a service. We can build on unified distributed storage system and
9:59
we can have various strengths like we can have local regional replication. We
10:07
can have the local replication of a data. If my data is at Regina that is in
10:14
North America, I can have three replicas, maybe in New York or maybe in that
10:21
local region itself. We can protect our data against this node-rec failures. We
10:28
don't have to worry about all that stuff, whereas global replicas system is we
10:34
have six replicas, two regions or three regions. For example, I want my data to
10:41
be stored in if my application or my business is at Canadian region and I want my data to be stored
10:51
in Europe because of because of I want different continent to mitigate the risk of disaster natural
11:01
disaster so that is the reason we have strength of GRS we can have our data replicate on any
11:09
location. Another benefit is if my data is at Canadian, if I have my business in
11:19
Canadian location or North American location and in future if my business is
11:24
more in Africa then I don't have to worry about it, we'll take care about that
11:29
It's protect against major regional disaster that is another requirement of business. Royal Bank of Canada can't afford to have their database in same
11:40
geographical region or any bank or any business can't afford in case of any
11:48
disaster if they lost data. What can be the repercussion? We can think about that
11:53
Another option is GRS plus read access to Scandry. We have Scandry
12:01
endpoints. So we talk about Azure storage. We can store it in local region, we can
12:09
store it into global region and we can have read access storage about, you know
12:15
on different regions with more strength. So it gives us more consistency
12:22
We always scrub our data to preserve from data integrity. So that's the
12:29
reason we are using Azure storage and now it's time to have you know one one
12:38
demo but prior to that I will cover another some basic points what has blob
12:45
storage object storage its store and some unstructured data unstructured data means app and web scale data big data from IOT Dix I was talking about
12:58
genomics etc. In genomics for one human being more than 13 million
13:03
required happened for that we definitely required unstructured data backup and archive for example a nuclear power station log file the data which is sent
13:17
by the satellite so definitely that is form of unstructured data and we don't
13:24
know what type of data how much it have to append so that is another feature we
13:29
have to we don't need to worry or we have strength to store data in PBs of
13:36
data it's durable we can use LRS and GRS global or locally replica system it is
13:42
highly available we can't afford or any business can't afford to lose data any
13:48
point of time or any downtime. So it has 99.99% of availability. It is
13:57
strongly consistent, dynamically scale up and bandwidth. It is cost-effective. We talk about cost effectiveness because it's pay per use. If you are using it
14:08
you are paying it It not about something that you have some plan where you are not using it you are going to pay So that is the Blob Storage Services which is in high demand because of data data integrity because of the
14:26
data required strength, the agility, scalability, security all that cover all
14:36
the things are covered by Azure Blob Storage which is unstructured data. For
14:42
For example YouTube images or videos that is a form of unstructured data
14:48
They are using storage. So that's the blob storage. Some basic concept which I touched through so that you have background before we dip
15:01
down into code or tools. But certainly I'm sure after this session you guys have a fair idea about blob storage
15:11
Why it is ideal for platform as a service? Why? Because it has limitless scale, globally accessible
15:19
I'm sitting over here, I'm on my vacations in Mexico and I have to change the GRS to LAS or change the security or change the container
15:35
I could do that from anywhere where I have internet access. It's not about I have to move to that data warehouse physically to make the
15:46
change. No, I can do it with tablet, I can do it with mobile if I have access to
15:52
internet. So that is the global accessibility. It's less dependent on a
15:58
physical location. It's cost-efficient. Scenarios for applications where we have to use it. We have live data repository, for example, we talk about say Chicago
16:12
traffic management. They are recording all the data of different streets of
16:18
Chicago to look over the security and traffic that is come under data
16:26
repository umbrella and that is unstructured data, active and deep archive. If we have to preserve the data for some nuclear test which is going on from last
16:38
12 years or some research then definitely storage blob storage will be
16:46
in use and data ytics. So what is Azure storage concept? Here is we we are
16:54
on Azure storage cloud we can have one person can have multiple subscriptions
17:01
and one subscription can have multiple account say if here is anybody of us we
17:12
can have multiple subscriptions subscription one two three and we can have various storage account what is the benefit what we will achieve we will
17:21
achieve we can keep the separation of concern we can have a different to
17:29
different storage account related to each other probably we are involving on
17:34
ten different activities so we can have ten storage account so it doesn't
17:38
restrict anybody to have one subscription versus one storage account Container is kind of folder, you could say our folder at system. Beneath that we
17:52
can have our data. In this case today we are talking about the blob data or blob
17:59
storage. So that's the reason. How does your storage work? If you have
18:06
to access if you want to share your data to me I'm sitting on different
18:13
geographical location I think everyone of us is on different geographical location how we could share the data we could share the data with the URL where
18:22
HTTP storage account would be the deeper classic for me or for you Carlos or
18:29
Dixar, Rahat anybody, blobcode.net, container name or blob name that is the
18:37
we can give the blob name we can have container we can have one container
18:42
under one container we can have multiple blobs so that is the storage URL
18:47
Type of blob, how many type of blob we have? We have three type of blobs one is
18:54
block block block block are you know we store the data in blocks that is more
19:01
commonly used 90% of our development scenario we are using the blocks a
19:08
storage for example documents we have one document for Regina chapter another
19:15
is for my Chicago tech talk another is for my Montreal tech talk so we have
19:21
different documents, images, we can have different images, video so maximum time
19:27
we have block blobs second time is append blob, append blob as name suggest
19:33
which is continuously increasing for example log files they are increasing day by day we can't break log files because of the consistency satellite is
19:45
sending the data and that is appending data, you know. So that is logging big data
19:53
that is append data. Page blob are kind of index on our book so we
20:02
name them okay 512 to any number so that is three type of blob. We look
20:10
our block block 90% of our time we will be around this append block can be when we are playing role
20:19
of like one of my friend who is working on international atomic research center he they
20:28
are using you know append blobs or page blobs as per business scenario for example banking system
20:35
probably they need it for particular brands so they have page blob. So block
20:42
names for example for us it may be the perk-blob.core-image. This is my
20:50
September agenda chapter. I present the same so yeah that's what can be
21:01
blob name for you it can be anything you can have your container name you can
21:08
have your name and there are certain rules we will look over when we will
21:13
create the blob when we you know do a demo account name can be 3 to 24
21:21
character lowercase only that why you couldn find DEE as capital because as it lower case only Container name is 3263 lower case only Blob name is 1024 case sensitive
21:35
And that's the some rules of blob names, blob structure, standard property
21:42
blob blob, page blob, append blob. We look over different these E tag and last modified
21:49
we have the blob where we maintain the concurrency when this blob has modified
21:58
or we look over the size in bytes about the blob. We can have up to 8 KB of unique name associated with it
22:10
And we can have up to 5 TB, 5 terabyte per blob
22:17
one blob we can have 5 terabytes. We can have n number of petabytes data that is not
22:26
you know restriction for us. So now it's time for us to have blob storage for a demo
22:37
and for demo I'm going to my azure portal let me open it my azure portal this is
22:54
my azure portal as Rahat has also told you if you don't have subscription in
23:01
Azure portal please create your subscription. Its subscription is $200 we get credited
23:15
So we can play around that in that amount. So we have some stories in stories for Blob Stories
23:24
We are doing one demo. Someone has just arrived. I want him to look over
23:34
What we are doing, we are creating demo for Blob storage. Here we are on storage and from storage we are creating the Blob file table queue
23:47
This is where we are creating our Blob storage. Let me give the name. We will give the webinar May 27th because everything
24:02
should be small. .co.window.net. Deployment model is the resource model or classic model. Whenever we have any doubt we can use this sign to
24:16
get the information. It's a resource manager for new application of maximum
24:23
time we are using it. Account type say blob storage is our account type, standard or
24:28
premium. For standard, we are maximum time using standard. The reason is
24:35
premium is bit expensive. If your business required very like you need very
24:43
good performance for example banking or financial transactions then probably you could use the premium. Replication, we look over three type of replication in
24:55
our presentation decks earlier. There are three types of replication we can say
25:02
locally redundant storage, geo-redundant storage, locally redundant storage say in in same geography, georedundant storage in different geography, read access
25:13
georedundant storage is you know with some extra features, extra strengths over there
25:21
So access tier, cool and hot. Cool when our data is not used frequently, we use cool
25:32
HOT if our data is used very frequently, we use HOT. So, for example, for finance transactions probably we would be on HOT, whereas for government website
25:46
where the number of users are not so high, then we could use COAL
25:53
Choosing an access tier, let you specify the access platform for data residing in Blob Storage
25:59
HOT access tier is idle for frequently accessed data. So when you create today this, definitely you can try to create different options
26:12
I'm using my MPN, Microsoft Partner Network subscription. You can have developer subscription or another subscription
26:22
Resource group, you can create existing Deepak resource for webinar, which we use for my webinars
26:31
Location we have I think I got on my website couple of questions about
26:37
location also. So we have different location I could choose Canada Central
26:45
this point of time. So we have created webinar May 27th Blob Storage which is
26:57
unstructured type of storage. It will create it. It has created it. It now
27:03
deployment in process. We have our deployment that's cool. Webinars, we use
27:13
standard, Canada Central, cool type and it's deploying it. Meanwhile it's deploying it I would also like to share if you guys want to do practical, want to
27:30
download code, this article is published on CsApp Corner. I am
27:40
sharing that URL at GoToMeeting also. You can have all the details from
27:49
there so that when you do if you want to increase your understanding or you want
28:00
to do practically implement it after a week or month or any time you will find
28:06
the details or you can visit deepercossit.com where you have you know
28:12
same resource same data over there you can send me questions and I'm happy to
28:19
help that. Meanwhile we were doing that I think we have created our first blob
28:28
storage and yes we are here
#Cloud Storage
#Computer Drives & Storage
#Data Management
#Enterprise Technology
#File Sharing & Hosting
#Programming
#Self Storage
#Shipping & Logistics
#Web Services