 Hello, welcome to SSUniTech Society side and this is continuation of SEO Databricks tutorial. So we are in the part 4 of this video series. So today we will be understanding about the Databricks architecture. So before going forward if you haven't watched the first 3 videos of this video series. So I would strongly recommend to watch those videos where I have discussed about the introduction of SEO Databricks and how we can log in inside the SEO Databricks service and why we should go with the SEO Databricks over SEO Data Factory. So if you want to understand then you can watch first 3 videos. So let's get started with today's video. So as a developer it's not very recommend to understand about the architecture of the Databricks but it would be great if you will be having the understanding about architecture of Databricks. So go to the next slide and we will try to understand each and every concept of the architecture. So basically inside the Databricks it is having 2 different pane. The first plane will be the control plane and second will be the data plane. And control plane will be our label on the Databricks subscription. So that will be your SEO subscription and data plane will be available on the customer subscription. So why it is available in the customer subscription because data is very precious and secure. So it will be available only on the customer side. So customer will be the owning this. So that's why it's available in the customer subscription. Now inside the control plane here it is holding the SEO Databricks UI and SEO Databricks cluster manager and it will also be having DBFS which is the Databricks file system and the cluster. So these things will be available on the control plane. I have discussed basic about all these things in our previous videos. So you can watch from there for the better understanding. You can only understand here like we will be having the cluster and this cluster will be created on the control plane and which will be on the Databricks subscription. Now go to the data plane. So inside the data plane it will be having the VNet and it will be also having as your blog stories and Databricks workspace. So while we are creating the Databricks behind the scene it will be creating 4 services. First one will be the VNet, second one will be the network security group, third one will be the SEO blog stories and fourth one will be the Databricks workspace. We are creating a single Databricks and under that these services will be created. Now here like your DBFS will be hosted on the SEO blog stories that we can see here. Next we will be trying to log in by using the Azure Active Directory and once we will be successfully log in here then we are required to provide some of the tasks. So those tasks will go inside the Databricks cluster manager and cluster manager will be going to go inside the Azure resource manager and resource manager will be taking care for distributing the task. So how it will be distributing? It will go on the VNet and under that it will be creating the virtual machines. So these virtual machines will be created based on the nodes of the cluster. So once it will be created the task will be distributed among all these virtual machines and once everything will be there it will return back to the Azure resource manager to cluster manager and back to the user. So simply you can understand you will be going to log in by using the Active Directory and you are asking some of the tasks like we want to fetch the data from that file. So we are asking for those tasks and here the Databricks cluster manager will take care for those tasks and will reach out to Azure resource manager and resource manager will go on the VNet and will be checking the virtual machines and those virtual machines if any one is available then that task will be there and if two are available then that task will be split out into two virtual machines and after getting the response from there it will provide the response to the cluster manager and cluster manager will be providing the response back to the user. So this is the overall architecture of the Databricks. So it is not very difficult it is very straightforward you can easily understand. I can understand like all these concepts are not very understandable right now but don't worry we will be discussing each and everything in upcoming videos in detail and with practical. So you will be understanding those and once you will be understanding those individual concepts then you can revisit on this video and then you can again relook on this. So thank you so much for watching this video see you in the next video.