 Welcome back, welcome back. This is still why in the morning and your own time for the first interview of the day for just doing us now we have come from a banter with Grace and now we have moved to sports on tech and today we're discussing something that might seem serious but it's not that serious. We're going to break it down in a very simple way so that you can understand. We're talking about cloud computing and for that we have been joined by Padi Adela who's a cloud architect to help us with the conversation. Can we stand up Padi? Thank you very much. Glad to have you with us. Lovely to be here. It was nice. Okay tell us for someone who's just hearing a cloud architect for the first time what does a cloud architect do? Happily. So I work as a cloud architect at Dimension Data and my role as a cloud architect is I listen to customers concerns or customer needs in regard to their need to move to the cloud. So what happens is I look at how their infrastructure or their services are currently set up. We draw up solutions for them, discuss this with their clients and finally migrate them to the cloud. So I'm a superhero for making sure clients are happy as they run their infrastructure and services in the cloud. Okay so it's more IT related? Yes it's an IT related role. Okay awesome. So now tell us now what is a cloud for those that also don't know about? Sure so the cloud I think is a name that people came up with because of the icon looks fancy and shows that you are somewhere that's not where on the ground but ideally cloud computing is the adoption of IT services that you consume that are not within your reach. However when you look at the type of cloud services we have then you realize that we have private cloud which is services that you run on premises that is in your data center. For example Y254 would have a data center somewhere to process your computing needs and then we have public cloud and this is the one people commonly refer to as the cloud which is services that you consume from cloud vendors AWS, Azure, Oracle and Google. You also have a hybrid cloud where you work or you rather you consume on-prem services together with cloud services and you have a community cloud finally which is what institutions like universities would have where you share resources from a common point we'd call that a community cloud. Okay so we have four types of clouds you've mentioned private, public, hybrid and community so we went to understand you at least you've given an example of the public cloud that AWS, Azure, Google cloud part of it now the private cloud. What is an example of a private cloud that we use here we can use here you know in Y254. So in Y254 I'm sure there's a room you call your server room right and in this server room you have computers or other end machines which handle the processing of the signals you generate how you propagate it and how your broadcasters view your services so that's ideally Y254's private cloud. Okay and the hybrid is a mix of both? Yes a hybrid now is because you have streaming services on the internet and the streaming services get the content from their on-premises devices then you have a hybrid model where you're using the internet services together with the services that you have on-premises. Okay and the communities where there's shared resources? Shared resources between organizations say the government the government's cloud is a is a community cloud because it's shared services between different agencies of the government. Okay so what type of cloud would a personal need to share a personal data in? Why would you want to share personal data? To save your personal data? To save personal data. So ideally you save your personal data in the public cloud and one of the services that people have largely consumed over a long time is storage services that's for Google storage services for Android users and iCloud for iPhone users. So those are public cloud services that we as individuals consume on a daily basis one for storage two for email services and three for communication services as well. Okay what's the difference between Google Drive and Google Cloud? What's the difference between Google Drive and Google Cloud? So Google Drive is a storage service. It's a drive where you store your photos, you store your files, so on and so forth and Google Drive has been there for a long time as a personalized service. However when you look at Google Cloud now as a cloud service it encapsulates much more than storage. The aspect of computing comes in where you can now run your computing services inside Google Cloud. It encapsulates data analytics where you can take advantage of data science tools within Google Cloud. So Google Drive is primarily a storage service but Google Cloud now comes with much more than just storage. You have computing, you have storage and you have data analytics to go with as well. Okay and there's this you've mentioned the different types of clouds. What's the comparison between an on-premise and cloud computing? Sure so let's start with how you get the infrastructure. When you want to set up an on-premises cloud service then first you have to scope for equipment. You have to know what computing needs does our organization have. This influences the type of devices or the power of the devices that you will buy. This will power requirements in terms of powering the equipment coming, cooling requirements coming, resources to manage these in terms of the IT guy, the storage guy and the power guy coming. So you see as you're setting up your on-premises computing services then you have a lot of gears moving individually to come together and make up the on premises cloud computing services. However or on the other hand when you look at public cloud services if we need computing services today for a public cloud all you have to do is create an account and allocate resources, spin up the instances that you'd like to consume and start moving. So that's one of the comparison points that you have when you look at on-prem versus cloud. There's a cost on-prem, an initial cost of procuring the devices that you will use. Whereas on cloud because it largely appears you go model and simply create your account and you start using it. Okay. Two for on-prem is you need to take care of your security. Your IT team are the ones who are going to do your patching. Your IT team are the ones who ensure that regular updates are applied. In the cloud model it is the cloud vendor who is responsible for patching of the software or the services that they offer you at a high level. When you go on-prem and you need to upgrade or say Y254 is the services or other the computing resources that you have are no longer enough for your requirements, you would need again to procure new devices and this comes with the lead time for shipping from the vendors. However in the cloud we have elasticity. So elasticity means that when you observe that it is not enough say on your phone you observe that it is slow you would need to buy a new phone right? Yeah. But in the cloud when you observe that it's not enough for you, you simply increase the capacity of what you've provisioned and you're good to go. So there you enjoy the benefit of elasticity. Okay so I'm getting there comparing the two sides for business to probably advise them to do the cloud model. So usually you need to look at your goal as a business first as you explore which computing services would be ideal for you. Because again this is the cost versus risk. So is it easier to set up a data center for you than move to the cloud? But that means that you lose out on elasticity compared to those who sit in the cloud. Is it more secure for you to run in the data center than run in the cloud? See in the cloud your services are away from you but within your data center if you do not need to access the internet then you'd happily run a secure system on premises. So you really need to consider your business drivers for adopting some of these services before you make a decision on them. Okay speaking about cost which one is more cost effective or does it also depend on the needs that you have? I'll give you an example of a startup right? You and I come up with a startup and we need computing services. For us to procure a server, a proper server, we'd need to project how much computing would we need over a period of time. We'd need to order it, have it shipped, install it and then bring up our services. Whereas in the cloud we'd need to simply create an account, provision our services and get moving. So in that case then it's easier for us to quickly adopt public cloud services so that you can start production. Okay. So cost again is influenced by your business needs. Okay. However in the cloud it appears you go model so you can always adjust your cost, you can always optimize your usage so that you can have better running costs in the public. Alright being that it appears you go model what happens if let's say that we're having some trouble in the company and I can't pay for like the next two months what what happens to my data? So we mentioned that it appears you go therefore you build at the end of the month right and depending on the cloud service provider they have different service agreements with those who consume their resources. So there's always a grace period that you're given and you're constantly rather continuously reminded about your pending bills. In the event that that grace period lapses the cloud service provider notifies you that your account is due for closure and your data will be lost. Because again it's a business. Okay. Alright. Explain the Lord. Now to the cloud deployment method models rather what are some of those? So deployment depends again it goes back to our public you can either deploy as a public cloud you can deploy as a private cloud that is on-prem can deploy a hybrid model and you can use a community. Oh so it's basically that. Yes. Alright I thought it was something different. Now I know also there's this thing that you you said you mentioned in the previous conversation that you are in about data security it being one of the concerns that there is with the technology you said data security is a shared responsibility between the provider and the user. Tell us a little bit about that. So I'd like to start off with an with an aviado for example when you buy a house you buy a house with locks and a security system however it is your responsibility to lock the house at night. Right. Yeah. The same applies to the cloud. So in the cloud security is a shared responsibility adventure between the cloud service provider and you as the user. The responsibility of the cloud service provider is to secure the infrastructure that they give you that is the hardware wherever it is set up and the network requirements ensuring that security patches are continuously applied to them. You as the user comes in with your data therefore it's your responsibility to ensure that access to data in the cloud is secure that is who can access your data the users within your organizations. How do they access this data is it using secure protocols to access it and are you logging how they access the data and what they do with this data so that in case of an incident you can always track and remediate. So again security in the cloud is a shared responsibility model where the service provider takes care of security of the cloud and that is the infrastructure and the underlying services whereas you is the user takes care of security in the cloud that is access to your data and how it is used. Okay. Makes a lot of sense. So you take care of access then they take care of the infrastructure. All right. It makes sense. What about data recovery? What happens? Do they lose data also and what happens if they do? Good question. Data recovery depends on how you set it up. So we usually have different data recovery models again for the cloud and this is initiated by you as the user because again when you decide that this data is not important after generation then you can as well let it go. However, if this data is important to you then you need to come up with proper backup and recovery methods for your data. So one is you could simply backup and restore whenever you need it. Two is you could always run your resources actively so that if one of your nodes fail then the other one is always up to recover. Okay. So depending on your needs for recovering your data or securing your data after use then you choose how your data is ideally backed up and how it is restored when you need it. All right. What about data privacy because I know we have a data privacy law in Kenya but does it also apply to international companies now that we're dealing with Amazon or dealing with Google? Does it apply in case of data breach? It does. One, as I mentioned earlier, security is a shared responsibility, right? Yeah. So whenever there's a breach from the cloud service provider it is their responsibility to remediate. Whenever there's a breach because you as the user neglected something you need to remediate. When it comes to data policy and data residence that's where a law comes in and the requirement from the office of the data protection office is as much as you're consuming public cloud services you need to have a copy of that data locally within the country. So as long as you meet the condition that you have a copy of the data you're running or processing in the cloud within the country then you're good to go and free to use cloud resources. And as an addition to that a lot of the services you consume locally especially in the FinTech industry is processed in the cloud. Many of them are using AWS. Many of them are using Azure because as long as they have met the requirement of the office of the data protection policy where they have local copies of the data they process in the cloud within the country then they're free to go and use next-gen technologies not limited by lead times and physical service. All right. I'm getting you there's no bad side to it from what you're saying what are some of the challenges that come with cloud computing? What are some of the challenges that come with cloud computing? So there's usually a joke about people setting up a system administrator who set up cloud services and an exorbitant bill comes in at the end of the month. What this means is you need to be proactive to control your billing and resource usage. And the cloud has come with solutions to this. You can schedule your computing needs so that at night they're shut off whereas on-prem you can't switch off a server and come put it back on in the morning. So in the cloud you can always control your billing. You can set up alerts and budgets. If you'd like to consume 500 dollars you'd put an alert at 400. And when that threshold is met you'd receive an email or a text to go and check into your resources. There's always the risk of because you're running in the internet you as a user doesn't properly secure your environment. Therefore you need to bring in security consultants to ensure that your ship is watertight and you don't have a hole because again lose, leaps, sink ships. Okay. All right. But is there also an aspect of lack of control to an extent because when you take in the cloud services your sort of relinquency the English relinquish some of the services to them such as the IT management. So is there an aspect of lack of control too? As a business if I can relinquish. Got it. IT services to focus on my core mission I would be happy because this means that if I'm a vendor and I can focus on my customers rather than the backend of my IT requirements it means that more time goes into my core business and I become more productive. I mentioned it is important though to make sure that all the I's are dotted and the T's are crossed in terms of control because if you absolve control then catastrophe looms. So as much as your relinquishing control I love how you've done it. Or by your IT services it means that if it's set up properly and you're not really giving up control it's just the mundane services that boggle your IT team every day such as updates patches provisioning of new resources and a lot of time spent in tasks that have been automated reinventing the wheel then you can now enjoy automation and the ease of service delivery focusing on your mission as a business. What about as we come to a close on the conversation where's the incorporation of AI, IoT and everything in cloud technology? There's a lot of it actually and I think for a lot of people who are pursuing rather a lot of people looking to consume AI services then the cloud is a really good place for you to start. For one there's a lot of community around AI who operate in the cloud because for us to set up AI in here we'd have to buy capable servers we'd have to bring in experts to set the baseline for our infrastructure and then start building services on it. However in the cloud AI is consumed as a service. Remember earlier on before they show you were talking about infrastructure as a service software as a service and platform as a service. So the cloud offers you a platform as a service where the baseline of the infrastructure has already been set up for you therefore you as a business say if I was a financial industry who would like to do analytics on who are our customers and the demographies around it or simply bring in the data, quickly set it up and generate insights from it. Therefore you can do a lot with analytics and data science in the cloud because the platform has already been built for you with analytics in mind. So it is bespoke for analytics rather than wanting to set up or inventing your way within your offices for it. Okay interesting. Finally now for businesses that are still operating in the manual methods, old methods of storing data of doing their services, how important is adoption of this technology? Adoption of cloud technology is a wave already and it's a tsunami that's sweeping across the lay of the land, carrying everyone around it. Therefore for businesses who would like to consume cloud services it is important not to jump the gun. You need to take into consideration your business needs first of all why, there's always, I've heard a lot of people talk about asking the question why. Why are we in need of consuming cloud services? The next thing is how then are we going to consume them and this is where Padi is a cloud architect who comes in to analyze how are you currently set up? What services are you running and what are the interdependencies between the applications. Once this happens then we model for you how this will sit in the cloud. Therefore we come up with architectural designs and we have continuous workshops with you so that we chart the way forward. This is now in preparation for moving. Then we start moving to the next workloads first so that you can get the gist of it and you can observe maybe you put in two workloads there and help breaks those and you have to take a step back to rethink it. But again as you test and it is a good practice to test in cloud, in software development, in the legs then you get the gist and you get to control the variables around it and finally once you are happy with that then now you can deploy applications in the cloud and consume cloud services giving your customers a happy experience. Important to note that if forgot is as much as we have a shared responsibility model, the weakest link is always you as the user. The weakest link is you who writes down your password. The weakest link is you who doesn't set up budgets therefore a lot of the breaches that we have observed is always a user who forgot to do something. So it is important again to sensitize all your users and all your resources within your organization about the importance of security and the importance of following protocols and policies that are set up for them. What are some of the companies that have adopted this technology? I know I said we are closing but you are still closing. What are some of the companies that have adopted this technologies? I am not at liberty to disclose whom have moved to their cloud because of NDS with them. However our government is a large user of cloud services and a citizen runs somewhere and is close but again you access it over the web therefore regardless of whether it is a public or private cloud then you are consuming cloud services. So as an end user you get to consume cloud services regardless of public or private but you are consuming cloud services. If you would like to know or if organizations would like to know the companies that they could emulate to move to the cloud then this information is readily available from the cloud service provider's websites. You can see a list of all the companies that are consuming and even during sporting events you get to see all these cloud service providers ads on and you see AWS all over the place. You see Oracle all over the place on the vehicles. So this shows you that a lot of huge companies are already consuming these services. Netflix is a huge user because you can see the deployment models around it. Therefore people are moving but again be careful about where you are moving and understand so that you can move properly. So you have knowledge even as you plan on migrating. This being your career, your position, it's not something people are used to hearing a cloud architect. What does it take to become one? Does it take to become a cloud architect? My background lies in telecommunications engineering. Therefore as a telecommunication engineer then you learn about internet protocols. So as a cloud architect one you need to understand cloud computing services, public cloud computing services. So you need to know what happens in AWS, what happens in Google and what happens in Azure. And the baseline model for all of these are all the same. It's just the services that they offer that are given different names. You need to understand networking so that if you're setting up Y254 in a hybrid model then how are we connecting the two? You need some Linux skills because a lot of the machines in the cloud run on Linux and there's a common saying that the cloud is a Linux playground. So what else do you need to understand? You need to understand communication skills. You need to be able to interact with your customers. You're not the legacy computer guru with a hoodie in the basement. You need to be able to articulate your ideas listen to what the client requires and draw up solutions for them. And I think all in all with the proper interest and a passion to understand the technology then you would become a proper cloud architect. I know cloud architects who have backgrounds in medicine so So it's possible for anyone? Yeah and the coolest thing about it is you don't have to go to hospital when you work with them. Ask them this idea. Therefore you just need to have a passion to understand the technologies. That is the core cloud technologies. You need to understand security. You need to understand how Linux systems work. You need to understand how networking works and you need to be able to articulate these ideas properly so that you can talk with your clients and draw up solutions for them. All right. Awesome. That's where we close it. Thank you very much. Great. It's lovely to be here and all is a pleasure talking about cloud technologies. I would highly encourage people who are thinking about abducting the cloud or becoming cloud architects to take it up. It's a fun field. It's an enjoyable field. There are sleepless nights when you drop solutions. And it's not for NADs only. But it is highly rewarding and it's not for NADs only. Okay. Thank you very much. You want to give you a social handle? Yeah. No. So I am not there anymore. I left. Okay. All right. So no problem. Thank you very much. That has been a conversation around cloud computing. I hope you've understood one thing or the other and if you're in this space well now you know what if you want to get into this space you know some of the things that are required of you. That's why we bring it to a close on Spot on Tech. But more is coming in entertainment with Gracia. My English will stick with us.