Cisco CCIE Security 350-701 – Network Automation Part 3

5. What can be Automated – PART 2

Apart from that, we can also automate the scheduled software deployments. So basically what is software deployment is like? Software deployment is nothing, but it’s a process of remotely installing the software on multiple devices. Like it can be computers, if you take an example of computers installing the operating system or generally when you talk about network devices, it can be router switches or firewalls where we are doing some kind of iOS upgrades.

Okay, so this can be done from a centralized location means you don’t need to go to each and every device to do that. I can do that scheduled, I can say schedule the upgrades of the devices and this upgradation can be done manually or automated manually. Again, you have to go to each and every device but I can say around 03:00 A. m. Probably identify the devices, let’s say if any router is running 300 900 series routers and if they’re running 50 iOS versions, identify those particular devices and upgrade them to 54 iOS version.

That’s it. So this process can be automated and it can be scheduled where we can deploy the software upgrades automatically from a single centralized location means you don’t need to go to each and every device to do this. So dynamically it is going to do this in the back end and this is going to save your time as well as the efforts. And apart from that, even though you did some scheduling, let’s say, you can also verify the status whether the particular task given or automated or scheduled is completed or not, because you need to make sure that it is also completed or upgraded. If it is not completed, then basically you can also verify what happened, why the process has been failed or what is the reason behind that. Even you can get those verification status information as well. Now apart from this, what are the other things can be automated like topology, visualizations and the documentation.

Like already said, the enormous scenarios you have to go to each and every, let’s say you have to go manually and you have to use some kind of software like PPT or some kind of video software where you need to make sure that this network diagram is generally created manually. And if you are making any changes to that then you have to go and make any change. You also need to make changes to the existing documentation as well as your diagrams. Now probably these diagrams can be maybe can be wrong or it can be non consistent, it may not be consistent again or maybe there may be some issues, maybe some not updated one. So these are some of the general issues.

So with the help of automation tools, what we can do is we can tell this tool, the software is going to dynamically discover the devices and automatically it’s going to update your topologies and it’s going to provide you some kinds of visualization. Like this is how the visual it looks how they are connected so the same exact visual look how they are connected means because normally the diagram what we use is manual and the topology is also something different because if these two doesn’t sync and if you are trying to understand your network then there is a problem. So depending upon the changes what you do in your network, the topology will be dynamically updated. Let’s say if I shut down the link or remove the link from here or maybe I remove this device so this will be updated dynamically so you can automate the network diagrams. The documentation is up to date with minimum human intervation so it means you don’t need to manually do these things so with the help of this automation software these things will be done automatically. So apart from that you can the most important thing is like troubleshooting the problems because as the business grows the scope of your network is going to be bigger. Like, let’s say you have some let’s say you have some hundred, 1000 clients and you’re connecting your network and maybe in the future as the business grows, the number of users or the number of clients connecting may increase.

So that will again lead to more traffic because the number of clients increases and you’re using the same network, the more traffic and that is going to impact the quality of the application also because when you have more clients automatically that will impact the existing applications. Applications may not work properly because the bandwidth requirements will increase, the application requirements will increase because of the number of users increasing. So probably your job is to troubleshoot. So probably what you are doing is you’ll be troubleshooting that problems and the troubleshooting time also will increase because the more problems you have, the more application, the more problems you have, the more time you spend on the troubleshooting also will increase. And also the team the It team required to manage that you may require more people to do that as well.

So what I want to do is I can simply automate this troubleshooting. So what I want is I want to tell the automation software to okay, if you see this kind of troubleshooting error then you just go and do this. So we can do something called automated troubleshooting where it is going to scan all the networks down to the layers like from your physical layer to your application level and it can identify the health status of individual devices and also it can collect the status of those particular devices like all the devices in the network. What is the status and what is the status of the connectivity, the secretization, those kind of things. Depending upon that it can use some kind of simple troubleshooting steps like you got an engineer, let’s say he faces this problem, some XYZ problem, then you can automate that particular problem to fix that particular problem you can tell simple steps with options.

Just like you have in the windows. You have network troubleshooting diagnostic tools, you have some kind of GUI. And depending upon the problem you select, you will have automatic steps in the next window. So this way you can minimize the reduced time because based on these automated questions in the back end, this will help you to minimize the time for searching for the root cause of the problem. So, based on the set of questions, probably you can automate some of the basic troubleshooting steps and most of the automation tools, most of the software, they do have something called artificial intelligence, as I said, and also some kind of machine learning features. So probably with the help of this, they can automate most of the troubleshooting tasks. Of course, not all the automation software support this, but most of the automation software, if they support these things, probably you can automate most of the troubleshooting tasks depending upon artificially.

6. Impact of Network Automation

So the next thing we’ll try to see what are the various benefits we get with the network automation. Because when you’re adding network automation in your network we will see how it’s going to impact your networks or the world of networks. Now, with the help of network automation the network engineers or the network admin role becomes much easier. So at the same time it will also provide you some other benefits in terms of company wise like reducing the overall cost or reduce the operational cost. We can say because with the help of automation options the network cost will automatically decrease because most of the network activities can be automated which will provide you a faster way of deploying the devices. Whether it is doing the configurations or whether you are making any whether you are doing some kind of troubleshooting or whether you are trying to manage the other devices or deploy any new devices.

So all these things automatically this task can be faster which will again reduce the operational cost and also the number of network engineers required to deploy or manage that particular network. In terms of network engineering, network engineers point of view. We can say it minimizes the job complexity because you’re not going to do each and every you don’t need to really memorize each and every show command to figure out what might be the problem. Or you don’t need to remember each and every command line interface when you want to do the configurations. And in terms of networking is again whatever the task, when you do it manually let’s say it takes a few hours and that can be done let’s say it takes six to 8 hours. That can be done in less than one or 2 hours or with a few people. Again also it will save your time as I said. And also it will eliminate the repetition of errors. Eliminate repetition means again the same things you don’t need to repeat and also minimizes the human errors. Less likely the human errors are possible.

Now, the next thing is like the other options, what you get here is conflation consistency. Now, conflation consistency means we will be having some kind of stable configurations or stable working configurations which are as close to as what you want to be implemented in your network. Okay? So this will ensure that you have the exact same configurations what you are expected to configure in your network or as per your company requirements. And when you have the stable configurations, again it will minimize the time for the troubleshooting because when you have correct configurations in place most likely there are very less troubleshooting you need to do because you already have all the correct configurations in place. And apart from that we also have a better network control.

So with the help of automation software or automation programs we will have a perfect view of your network means kind of you can see the diagram how the devices are connected. You can see the status of your devices or the Cputilization or the number of devices you have, how many active devices are there and these all information. You can have a better view in a graphical representation which will provide you the perfect view of your network and you also have an access to your network from one single centralized place that is from controller you can access all your network, even your data and also all the performance reports easily. So this will improve your control on your network.

Now, apart from that, there is one more thing you get is elastic scaling. Now, elastic scaling is a term used to define kind of ability. We can say it’s a kind of ability to automatically add or remove computers, servers or networking devices. Like this was introduced initially like with the VMware service. Let’s say you are using some kind of VMs virtual machines where you have some kind of dedicated service to be installed, and the requirement of this service may change. Maybe, as per the requirement for this month, you require at least five servers to handle that load or the request. And maybe the next month you just need only three servers. Or maybe the next month you need eight servers. So depending upon the requirement, depending upon the requirement the infrastructure can be automatically added or removed completely based on the requirement like the number of VM servers which are required can be allocated or it can be removed as per the requirement. Again removing whenever we don’t require them. So scaling this can also be you can compare with the kind of scaling with van circuits. Let’s say you got multiple van circuits in your network and you’re running multiple van connections and probably the router may use this van circuit to forward the traffic.

So whenever it realizes that there is not enough bandwidth available on this link it can immediately start using the second link which is actually used as a backup and again it can remove that forwarding from the forwarding whenever it is not required. So this is also one example of elastic scaling where you can have a backup band circuit and that will be automatically added or removed or automatically used or not used depending upon the requirement. Now there is one more thing, just a small note you need to know specifically with networks so most of the network administrators most of the time what they do is they work with the command line interface. Like if you take an example of the network engineers working on a Cisco devices. So probably they will be doing most of the things from the command line where they will be doing some kind of configuration changes or verifying some troubleshooting with the help of some commands or doing some kind of monitoring with some kind of SNMP tools.

So this is something what is generally restricted the job of a network engineers associated with network devices most of the time and as a part of network automation now because the network automation is more and more common now. Mostly. Now, the network engineers need to have some little bit knowledge, or they need to be a little bit familiar with the server side operations. Like server side options. Like because all the automation tools, software, whatever you run, they run on specific servers.

And when you are running this automation software on a specific server, basically you need to know some kind of terminals like the CPU. Generally you have two CPU, two core CPU, something like that, three core, four core like that. And then you need to understand the terminals like core again, what is core inside and then software packages required or installation of your operating system, maybe an operating system or patches, these kind of things.

• Category: Uncategorized