 Hello, welcome to SSUnitech. So, we'll decide and this is continuation of SEO Databricks tutorial. So, today we are going to see the configurations inside the clusters. In the last video of this video series, we have seen how we can create the clusters and what are the terminologies that are available by creating the clusters we'll be seeing in detail in this video. So, let's get started. So, inside the cluster configuration, the first thing, which is the cluster mode. So, cluster mode is having two types. First is the multi node and second is the single node. So, let's discuss about the multi node first. So, inside the multi node, it will be having the driver node as well as the worker node. So, whenever we receive any request, so driver node will be going to distribute the task into the worker node. So, the task will be going to parallelly divide into the multiple worker nodes and worker nodes can be created as per the need. So, virtual machine or the driver node will be taking care for all these. Next is the single node. So, single node will be having only a single virtual machine and that will be the either we can call this as the driver node or we can call this as the worker node. So, whenever we will receive any request, so it will be receiving as a driver node and after that working on that as a worker node. So, it will be only having only a single node. Next is the access mode. So, inside the cluster it is having three different access modes. The first is the single user, second is the shared user, third is the no isolation shared. So, for the single user it will be only one user can have the access and it is supporting all four languages which is the Python, SQL, Scala and R. Next is the shared user. So, inside the shared user here the multiple user can access and it will be only available in the premium workspace and it is supporting the Python, SQL, Scala and R. The last one which is the no isolation shared. So, it means the multiple user can access and it is supporting all four languages which is the Python, SQL, Scala and R. So, this is the access mode available inside the clusters. Next is the Databricks runtime. So, Databricks runtime is having four different types. So, first is the Databricks runtime. Second is the Databricks runtime ML. Next is the Databricks runtime Geonomic and last one Databricks runtime Lite. Inside the Databricks runtime it is having the Spark and Scala, Python, SQL, R, Ubuntu libraries, GPU libraries, Delta Lake. So, all these are available in the Databricks runtime. For the ML everything from the Databricks runtime which is the first one and additionally it will be having some of the popular libraries which for the ML inside the second Databricks runtime. Third one for the Databricks runtime Geonomics. So, everything from the Databricks runtime which is the first one and additionally it is having few of the open source for the Geonomic libraries that we can see and it is also having few of the Geonomics pipelines and last one for the Databricks Lite. So, runtime options only for the job not required advanced feature. So, this runtime option only for the jobs. So, this is the Databricks runtime. Next is the Auto termination. So, auto terminations if you remember in the last video we have seen it was having 120 minute but we can change that if you are not doing any activity and your cluster is not going to use for few minutes or whatever the minute you have set it. So, it will be terminated after that. Next is the auto scaling. So, auto scaling option is very important. So, user specifies the minimum and maximum work nodes and auto scaling between the minimum and maximum based on the workloads. If you are having the maximum workloads so, it will be going to utilize more worker nodes. Here not recommended for the streaming workloads. So, if you are going to deal with the live streaming data then it is not recommended to all the auto scaling. Last cluster VM type size. So, here it is having total five different types that we can select first the memory optimized compute optimized storage optimized general purpose and GPU accelerate. So, accordingly inside that we can see it is having the different nodes different memory and those we will see in the practical. Let me quickly go and try to create another cluster. So, it will be very useful and easy to understand. Let me click on this create and in the last video we have created cluster. Let me call this for the testing multi node. Here the policy the policy we have discussed. So, it is having like total four options. So, I am going to use the first one and this will be available only for the premium workspace. Here the multi node and the single node so that we have already discussed. So, for the multi node here it is having the access mode. So, inside the access mode either we can go with the single user or we can go with the shared or we can go with the no isolation shared. So, that we have already discussed. So, let me go with the no isolation shared and after that here we can see the Databricks runtime version. So, we can choose the latest version always here. Let me quickly go in the downside and here we can see the photon acceleration. So, I am not going to deal with very last data. So, I am not going to enable this option. After that we can see the worker type. So, inside the worker type it is having the general purpose, general purpose HDD, general purpose delta cast accelerated. So, all these are here. So, by default I am going to use the minimum one. So, it will not be casting very much. After that we can see the minimum worker and maximum worker. So, minimum worker will be one. We can set it and maximum we can set as per your need. So, I am going to make this as four maybe for the maximum worker. Now, in the downside we can see the driver type. So, driver type we can same as the worker we can set it. Here we can see the enable auto scaling and the terminate after. So, inside the terminate after let me have this as 10 minute only. Inside the tags you can add the tags and inside the advance option we can see the spark and logging. Spark we can set the spark configuration and the environment variables. So, let me quickly go and try to create this cluster. So, once we click on this create cluster. So, it will be adding. It will take around five minutes to create this cluster. In between let me tell you about the configuration is here. Then the notebook. So, as of now we have not attached any notebook here. So, that is why it is blank. Inside the libraries we have not installed any libraries. So, that we can install if you want. Inside the metric we can check about the performance here. So, it will be having all the things of your cluster. So, like the nodes how many nodes are utilized, how many nodes are vacant. So, everything you can check it from here. Let me go into the configuration again and let's wait until this will not be created. Now we can see the cluster is created. So, it is taking around four to five minutes. So, you can wait while you are trying to create this cluster. So, thank you so much for watching this video. If you like this video please subscribe our channel to get many more videos. See you in the next video.