78. Around the Corner Agile, Cloud-Scale Applications, and DevOps
This is the last topic in this particular section. Around the corner, Zale cloudscape application and devil. This is nothing, but you can see the waterfall model that we studied in our, say, engineering career or maybe in BCA. We have this particular model where we start with analysis, planning, data, design, building, testing, and operation. Inside the operation, we have the same lifecycle. So whenever a particular project is going live before that, what are the things that we are doing? So, what we are doing, we are going to the customer providers, we are analyzing their requirement according to their requirement, what will be the It infrastructure we needed, means how much applications they have, how much storage they need, what is the network, services, all sort of things. And plus, by the end of the day, what is the uptime and what is the service?
What we can say that what is the operational things they are looking for. Operational things mean what type of service level agreement I have with them. In terms of, suppose if some network down is there or other stuff. So, let me make this easy. So, service level agreement is nothing but according to priorities or severities a p one to p four, how our technical team will react to those things once it will come into the operation. Okay? But before that, obviously, we need to analyze, plan, and design. This will again require so many things. While we will design then how to build. And all these models should be recursive in nature. That means that, say, once you design, it doesn’t mean that again you will not come back and redesign. So, if any requirement is there, or maybe you have a design failure, then you have to come back. And then again you have to plan and design and then build and test and then operate these lifecycles are iterative means it will go and you’ll come back and it will come back. Let me show you this particular process. So here you can see that inside the project timeline.
Once I start with analysis and planning, you can see design, construction, and testing. This is cyclic. Means once you design, built, test, again you have to design, build, test. Means if you have any flaws or if any other requirement come during the design phase, or during the builder test phase, you need to change that. Once you change again, it will go into the cycle. Okay? Then second you can see that customer turnover. Now, suppose you had designed a network, but due to some reason, the customer network has expanded again, so you need all those things. Like before, you have to go inside the lifecycle, analyze, plan, design, build, test, and then operate according to the given SLA parameters. Then if there is some technology upgradation or innovation. So suppose Cisco from Nexus Seven K moves to Nexus Nine K or some other automation or SD product they introduce into the market, then your design should accommodate those changes, which means your design should be in such a fashion that it can accommodate customer turnover or customer requirements. They can accommodate technological innovations as well.
Okay? These things are the requirements here. You can just stop or pause the recording. Just stop, pause the recording and you can just check that what is the phases and what are the definitions for, analyze, plan, design, build, test and operate. Okay? Now again, finally we come to the conclusion that your design or your phases are in such a way that they can accommodate customer requirements as and when they change. My design should be like that. That again it can do the implement and operate correct. So what is the requirement here? It’s very important to understand that’s why this Cisco devices, they are trending in the market because these devices has a huge capability. Even if you see a normal Cisco catalyst switch, it can do a number of tasks.
Okay? So that means that devices are capable enough that if you expand the network, it can expand. Now, for example, when it comes to technology innovation, say Cisco has iOS XE in some of the routers. This iOS XE is itself a big platform. That suppose if there is technology innovation, it can accommodate all those innovations inside it. Okay? Now we are discussing this with respect to clouds. Now where all these things will fit inside the cloud. Remember, I told you that inside the cloud you also have the data center, and these rules and principles can apply inside the data center of a particular cloud as well. And even there, they are using the same type of waterfall model inside the cloud as well.
79. Chapter 02 Service Providers and Information Technology
Hi everyone. Welcome to chapter number two. In this particular chapter, we’ll learn about a service provider. What is a service level agreement with cloud providers? We have various examples of cloud say infrastructures a service or platform as a service or say software as a service. So these type of cloud applications or these this type of infrastructure within the cloud for our premises we can use how we can use all these things we’ll learn in this particular chapter number two or in this particular section. So let us start with the first thing, which is the service provider and information technology. Now, when it comes to service providers, if you look at your organization, you will find N numbers of service providers and N numbers of vendors from where we are taking all these services. Now these vendors including say from a toilet paper to something like we are getting some sort of telecommunication service provider or network service provider means for each and everything inside our premises we use to take services and for that we have vendors.
So if you see here in this particular example, you’ll find that, okay, for telecommunications, for telecommunication, I have telecommunication providers such as AT&T. Verizon says that for storage, we have some providers, something like EMC or NetApp. For computer service providers, we may have Dell, HP, or HCL. For that, we have an Internet service provider. So whatever local internet service provider in that area even we can include say at and T or some other service provider vendors as well in this then we have primary and secondary data center service provider. Good example, maybe IBM or HCL or another big companies who are supporting the compute environment within the infrastructure managed service. Again, we can think this as an entity or maybe IBM or HCL, Vapor, they are providing this managed service infrastructure or maybe remote service infrastructure. So what is happening? Generally, the big companies, they run their application and for operation or you can say the operational things, they are giving that services to some third party. Okay? So that third party will manage all those network infrastructures, all those storage infrastructures, everything. So this will be the first layer before they meet the actual vendor. So what is happening? Let me draw here in the layered approach. So suppose this is the organization.
Now that this organization has deployed the services for different services, they have hired a managed service enterprise network. So they have some contact with, say, MSN, and they acquire that service. Now, what happened if any problem with this organization directly, MSN team, that service provider whose managing the services, they are interacting. Now if they are capable enough to resolve the problem, it’s okay. Otherwise, if they are not, then they are opening a tag case or vendor-specific case with Cisco, say, for storage, and with NetApp, say, for compute. If apps from IBM vendor, they can open a case. Sorry, IBM. So for networking, Cisco for storage, some other vendor for the data center—they have some other vendors like that. Okay, here you can check the details about all these vendors. So what you can do, you can just pause the recording, and you can read through all these things, and you’ll find it very much familiar. Because in day-to-day life, say, infrastructure, structure, and life, we find all these terms. Okay, so let me stop here, and in the next section, we’ll discuss the service level agreement.
80. Service-Level Agreement
Next, we have a service level agreement. Service level agreement is nothing but agreement between you and service provider upon some certain criteria. So what are those criteria? Those criteria are performance, uptime, MTTR (mean time to recover), and customer data handling. What does it mean by “performance”? Performance is nothing but the number of operations that the service provider must guarantee in a given time interval. That is very important in time interval with the offered capacity. These two things are very important when we are talking about the performance. So performance is nothing but the operation with respect to time interval and capacity. Okay, next is the uptime. Generally we are talking that my ISP has uptime say 99.99% or maybe triple nine or 59.So uptime is nothing but a measure of the amount of time a system must work correctly. So how long your It systems are working properly, that is the uptime.
So suppose some of the services are hosted by the service provider; how long can that be hosted? There should be minimum downtime, or if you want good uptime, you may have a primary and backup data center or primary and backup systems. Okay, then third, we have MTTR. In the meantime, what is the time frame in which that particular field system will recover? That is mean by the mean time to recover. Finally, customer data handling This is again very important because there should be some rule or some strategy upon which the service provider will base its agreement in terms of data confidentiality and backup. So your CIA confidentiality, integrity, authenticity should be maintained and obviously your deletion policies or your data privacy, all these things will be maintained. Okay, so these are the key terms for Sala performance: uptime, MTTR, and customer data handling. Next in this, we have the cloud provider.
So we have already discussed that if you are asking or if you are raising some request from the cloud, inside the cloud you have different different services offered. So we have a storage; obviously, we know that storage is something related to, say, a storage area network or a storage where clients can store their databases inside the server. Whenever we have generally we are telling this server to a storage communication. So whenever you have server to a storage communication, your servers can save the data and generally what servers saving the data in their storage those are the client request information or client information. So client information is first saved inside the server, and then it can be replicated inside your storage, if your storage has the saving capabilities inside the storage. Both things can work. It depends on what type of storage you have in your infrastructure.
But when we are talking about the cloud so directly from server I can store my data inside the cloud storage, then obviously we have networking. They are handling the networking task. Later in this course, we’ll learn about, say, CSR or some virtual routers. On that, we’ll see how they can do this networking over these virtual routers or over these virtual devices. Then they can provide the desktop services, the streaming, some audio and video capabilities, some web services, host applications, provide application services, and provide database services. Now, these three things are quite new: collaboration tools, publishing, and middleware. So let me show you the definition of all three. Here in this particular PowerPoint, you will find the definitions of all these terms. So when we are talking about middleware, it’s supplementary software that includes libraries, programming, language interpreters, database services, user authentication services, account management, and so forth. Okay? So it’s in the middle, providing libraries and binaries for different applications. Then coming to collaboration. As the name suggests, collaboration is nothing but collective effort, you can say or the joint work from different different people. One of the good examples is SharePoint, from Microsoft. SharePoint is one of the collaboration tools where different people can work and share their Excel awards or other documents. Inside SharePoint is the publishing application that facilitates the publication of text, such as a blog.
So Blog is one of the example. Even companies, they have their own publishing tool, they have their own actually customized publishing tool, collaboration tools. All they are using in one say some of the companies, they are using Yammer as a publishing tool. So different members in a team, they can create their Yammer group and they can publish their videos, audios, writing notes, all these things. Okay? And then finally, we have streaming. That is, again, I already told you that is the audio video features and the capabilities. So we are really good. Let’s stop here.
81. Infrastructure as a Service
Let us talk about infrastructure as a service, or IaaS. You can use this pizza as a service, as an analogy that in the traditional network, what do you have in the IAS, what you have in platform as service, what you have in software as a service, what you have but in coming two to three lectures, we’ll discuss more about all these analogy. Let us start with infrastructure as a service. As you can see here, this blue one is the provider’s responsibility. So what the provider will provide you provider will provide you virtualization server storage, networking. And you can think like this: okay, you have a virtual server, you have virtual storage, you have a virtual network, but you are not bothered about what that particular service provider is providing to you, then what is your responsibility?
Your responsibility is to run the application over your operating system and over your infrastructure software. Okay? So here I can see that some portions the service provider is providing and basically these are the compute thing like server storage and network. And for things related to the operating system and the application, you are responsible. One of the major player in this particular domains everyone knows that that is the Amazon Andes is one of the major player in this. You can see here in the Gartner coordinates that AWS is leading, although Microsoft is also following up AWS. But the market share is too high oafs with respect to any other vendors. Now, if we talk about this AWS, initially we have talked about the broad network access or network availability. You can see here that AWS has visibility across the globe. And in AWS terms, you can see here that they have regions and zones. So inside the region, they have Zone 1 and Zone 2. Like that.
This zone can be thought of as a data center. So suppose I’m talking about particular region, say US region. Suppose in US region I have for example zone one and zone two. That means in the US region, I have two data centers. One can work as a primary, and one can work as a secondary. Suppose if the customer is here, say this is the customer and it’s near to Saint, then it will not go to any Europe or US server for their fulfillment. Because in doing that it’s latency due to delay, everything will increase. So if it is near India, it will query an Indian server and get the response. Okay? So that’s why we need some sort of global ability, or in cloud terms, broad network access. Now, if you do the AWS course, you’ll find that they have more than 1,000 services or applications offered inside the cloud. And these are the, you can say the number of services they are providing. But inside the game development, again you will get some sort of applications and things. Okay? So let me show you quickly in this particular slide: What are the main services they are offering? So they are offering services related to storage. Let me start from the beginning.
So first of all, they have something called AWS global infrastructure. On top of that, we have all the services. So on top of that, we have computing, databases, mobile services, management tools for the visibility, messaging tools, developer tools, security, storage, blah, all types of things. the new things they have added artificial intelligence, internet of things. At the moment in present this artificial intelligence is trending very high and you can think that next it market in coming say three to ten years is purely inside artificial intelligence and at the moment this Internet of things is also peaking very high. So we have current business or marketed for Internet of things as well. These are the technologies that are trending at the moment. Okay? So inside all these things say if we are talking about compute, you can see here inside Compute we have services like EC two, easy to container service, light install number of services will find say inside storage we have simple storage solution like s three then elastic file system glaze here, storage gateway in database. We have relational database management, social networking, and content delivery. We have virtual private cloud, cloud front, route 53,all sorts of applications we have inside AWS here have listed few of the services and the definition you can have a look on that. Say EC two. So what EC Two will provide? EC Two will provide you cloud service that provide virtual services that are fully controlled bylaws user resize according to requirement of compute demand because obviously this is what Compute Cutco is elastic Compute. Okay, and it offers all sorts of operating systems. ECU Two enables robustness through its reasons and the ability to jun.
So remember, we have the reason, we have the availability zone, which you can correlate with the data center, and then we are providing security things related to the cloud that are interacting with the customer, such as VPN and other security protocols supported, and then we have the storage. So if you want to remember all these things, because there are quite a bit in number, so you can think, like to run a data center, what you need, you need network. You need a storage solution. You need management for visibility. So you need a management tool. You obviously need compute, but you need compute with respect to all of these. Obviously, you need security. So with respect to all these things, you will find the applications inside AWS. So some of the applications are related to storage (EBS, EFS), and some of them are related to networking, that is, the virtual private cloud. It is actually for networking, because generally, AWS is a public cloud, but inside that public cloud, you can mix in your private infrastructure. So at that time, this term will make AWS a private cloud. Later, we learn about the features of public and private clouds.
Then it will be easier for you to understand. There are some other features as well, like direct connect. Suppose if you want direct tunnel from yours to the AWS with this feature you can have that. Then we have elastic load balancing. We know the importance of load balancing. Even now, we have a dedicated chapter for load balancing. We’ll see how load balancing supports the balance between the user request and the available servers inside the data center. And then finally, we have Route 53. Say this: 53 is nothing but a DNS port number. So from there they had derived this route 53.And obviously, this is the DNS service inside AWS. These are a few of the services inside AWS. But you can see here that we have so many features in this particular AWS, and it is one of the examples of infrastructure as a service.
82. Platform as a Service
Next we have platform as a service. Here you can see in the diagram that, inside infrastructure as a service, the provider’s responsibility was to provide you with up to this layer. But in the case of platform as a service, you can see that the operating system and infrastructure (structure and software) will be provided by the provider. Now, what is the responsibility of the consumer so that they can run their application on top of the infrastructure provided by the provider? Now, it’s very important here to understand that this type of cloud or this type of offering is done from Microsoft, one of the leading inside platform as a service provider. And here also you can see that they have this much of global presence. Okay? So across the globe, they have their data centers. From there, they can provide this sort of platform as a service. Services, it’s very easy.
You can go and you can create your trial version just to verify that what type of services there in their Microsoft Azure portal. You can go and log to httpsportalazure.com you can create your account. Once you create account then so the thing is that go to Azure.com. So the thing is that, first of all, if you log on to Azure.com, it will redirect you to this portal. Hezu.com. Anyway, once you create your account, you will get a nice tour. It will show you that. Okay, do you want to create your resources from here? You can create the resources, and then you have the list of services. The top services are here, but you have the list of services. Now coming to these services. You can see here we have services related to database, we have services related to virtual machine virtual machines. That means we have supported operating systems such as Linux, Windows, and other operating systems as well. You can verify that.
Then you can see that we have load balances, we have a storage, we have virtual network, we have active directory monitoring, advisory security. So whatever things are applicable for infrastructure as a service, you will get all sorts of information or services inside this portal. Further you can move, you can have this nice search bar where you can search whatever you want to search. Then finally you have this notification, you have help plus support, you have a directory and subscription and then about the account information, okay? So very much it’s something like if you know about that technology, you have to click and you have to use that service. That’s it. Because everything is provided by the service provider, what you are doing is hosting your application, and that is the use case of platform as a service.