 book I recommend you to read it. It's the innovator method. Check it out. You would like it because it fits the concept that you're trying to do. What's the best approach that innovators use to implement projects like yours? The professor who wrote it is actually a Stanford professor I'm an interview with so many CEOs and teams all in the Silicon Valley and also Silicon Valley. And he came up with this new way of innovation. I have a book called the innovator DNA. The second name is the innovator method. Interesting. His name is Professor Furr. I don't know. Nathan Furr, his name is actually a very young guy. He's not old but I think you will definitely, I think this method will start replacing the normal management method, manage strategy plan and so on. And he's all up for experimenting. He's saying we're living in a place where there is so many uncertainties. You cannot plan for uncertainty but what you can do is design experiments till you find the right solution. Exactly. I'm looking forward to today's presentation. Looking forward to chat with you in person after this. Are you guys, are you guys still, are you guys going to university or still virtual? Everything is virtual. Well, my teaching is still virtual but I'm coming to the university just to feel the work, the whole balance. Did you get vaccinated, doctor? I am officially vaccinated. Officially grown? Hassan, did you manage to get vaccinated? Oh, that's good. Well done. Oh, it's really interesting. The second dose, different people have different experiences. Anya was, the second dose killed me. You felt it, huh? Yeah, I felt it so bad for two to three days. Two days I felt it bad and then from a sports perspective because I play pedal tennis so I felt it for two weeks. In my game, I felt it to us so bad. Oh, wow. Like slow reaction. Yes, and because of the pain in the kettle, low reaction, just slow and then muscles wise, you know, after an hour, you'll start feeling that you cannot, you know, fitness, you feel hit by it. Right. Nice. And then suddenly, I don't know how it's back to normal. It's like a switch. One day you feel very slow, the second day is like, oh, it's so bad. All of you living in a very weird world. Maybe an engineered virus. So when we start, we will have a button to share, yeah? Because yeah, now I can even share. Yeah, I, yes. I'm not sure if the participant, they can see us yet or shows live recording. I think you're live maybe. Oh, so they can see us? Okay. Nothing confidential. I think someone is asking in the Q&A from the participant, which vaccine you have taken. It's amazing. Just to test, do you see the presenter view or the, or the, or the normal view? I see the presenter view. And now? Now the, now the right, the right view. Okay. Okay. I just wanted to test since we have not started yet. Yeah, it's good now. Okay. We don't know. Yeah. And I think most probably not this week. Yeah. Okay. Okay. I don't know. I don't know. I will let you know later. I'm not one of the people who speak. I will let you know later. Salam alaikum. Wa alaikum salam. Can hear you, Raman. Okay. Sorry, I just had some technical problems in the morning. I didn't know that one. Yes. I let me just, we have 74 participants and good morning to everyone. Dr. Munir, if you would like to change your backdrop, please. Your background, sorry. Yeah. To be honest, I don't know how to do this. But okay, then no worries, Dr. Munir. No worries at all. So we have 72 attendees. We should wait just five more minutes until we have everyone attending. I know that all the speakers with us. We already 10 minutes late. So salam alaikum wa rahmatullahi wa barakatuh. Good morning, everyone. And thank you for your time in attending our digital forum on the digitalization of SMEs and alternative ways to survive. I would like to welcome everyone. This is a digital forum in collaboration between the Ministry of Transport and Communications. And I would like to introduce our director, Mr. Riaz Al-Safi, the director of the digital industry in the Ministry of Transport and Communications with pleasure to share with us his speech. Thank you, Riaz Al-Rawan. In the name of Allah, the Gracious, the Merciful. Greetings to our small and medium businesses, entrepreneurs, our partners from Hamid bin Khalifa University and all distinguished guests. I would like to welcome you to our first digital forum for 2021 as a collaboration between HBKU's College of Science and Engineering and MOTC's digital society sector. Our focus in this forum will be on offering Qatar small and medium businesses the opportunity to expand their knowledge and learn from experts on how to use emerging technology concepts in their digital transformation journey. Last year was a challenging year for all of us with COVID-19 being declared as pandemic and creating a lot of strain on our society and economy. On the contrary, it demonstrated to us that there is always a room for improvement in the way we work, live and connect with each other. Since the beginning of the pandemic, technology have given us the ability to stay in touch with our loved ones, the agility to continue working and find solutions for our most pressing issues and finally, the hope of returning to some version of normal life. In order to help our SMEs in tackling the impact of the pandemic, MOTC organized a series of interactive virtual workshops with 67 multinational, regional and local RCT service providers to help this vital sector, the backbone of our economy in delivering their services over the internet and doing it virtually and doing it effectively with the help of digital and RCT technology during which 3,895 SMEs attended and about 1,600 acquired new technology solution to kickstart or continue with their digital transformation roadmap. We also organized four digital forums in partnership with many distinguished organizations focusing on various technologies such as AI which helped our SMEs in understanding their importance in delivering new services, achieving economies of scope or scale or reinventing their operating model. In 2021, we will continue using our digital forums as a platform for helping our SMEs in their COVID-19 post-recovering by supporting them in innovating and building resilient and agile businesses. This will be done through raising awareness on the latest technology trends and concepts and engaging the various digital industry stakeholders and including policymakers, large, small and medium businesses and academia in one dialogue with the aim of accelerating digital transformation in the SME sector. We are excited today about today's topic, cloud services, which is giving businesses the ability to ramp up or scale down quickly without making any risky investment. This flexibility is vital for adapting to the uncertainty associated with the pandemic and also delivering financial savings when operating on an attack budget as businesses only pay for the resources and avoiding unnecessarily business downtime. While tomorrow's topic will be focused on how blockchain can help SMEs in building smarter and more secure supply chain, this includes its ability in tracing each transaction from point of origin to point of sale through transparent and traceable audit trail in real time. On the behalf of the Digital Industry Development Department, we would like to thank HBKU's College of Science and Engineering for their partnership and efforts in supporting our SMEs in their digital transformation journey. We also acknowledge their important role in having an innovative, digitally transformed, small and medium businesses in Qatar and looking forward to build on this collaboration further. Thank you. Thank you very much, Yazan, for your speech. And now I would like to welcome Dr. Munir Hamdi, the Dean of College of Science and Engineering from Hamid bin Khalifa University. Thank you very much, Rawan. And I want to welcome all the participants to this very important event. As we all know, I mean, nowadays we are witnessing what is called the Fourth Industrial Revolution, which is mainly focused and surrounded by data and data analytics in general. So the topics that are being covered in these two days are very timely and extremely important in terms of cloud computing and in terms of blockchain. In order for SMEs or even big companies to survive, they have to be at the forefront of these latest technologies. And we are honored to be working with MOTC, the College of Science and Engineering at Hamid bin Khalifa University. We have a lot of expertise in these areas and others. So we have Dr. Aiman Erbid, who's one of the experts in cloud computing, and he will be given a lot of valuable information to all the participants. The same thing with Dr. Muhammad Abdullah, who's also an expert in cloud computing, but also on blockchain. And tomorrow we'll have also other faculty members. Also, I need to mention the involvement and the expertise of Mr. Al-Hasan. He's our own PhD student, but also he's the founder of App Lab, which is a major software and cloud computing companies. We Qatar and all of us should be very proud of what he has achieved. And I'm sure all the participants will benefit from his expertise and from the services that his company can provide. I really think this is a very timely topic, and I want to thank MOTC for arranging this to all participants. And we will be looking forward to our involvement in these topics in these couple of days and future topics later on. So thank you very much, and I hope that you will benefit a lot from what will be presented in these couple of days. Thank you very much. Thank you, Dr. Munir again, and it is a pleasure always collaborating with Hamid bin Khalifa University. It's a combined effort, and we always benefit out of this partnership and support the SMEs within the journey of digital transformation. Thank you so much. Thank you. Please allow me for now to introduce Dr. Aiman Erbid from Hamid bin Khalifa University to cover his session on cloud services. Dr. Aiman. Yes, thank you, Rohan. Just a second to make sure we have the correct view. Yes, thank you. Thank you so much, Rohan, for the introduction. So today I will talk about digital transformation through cloud and edge computing. I'm Aiman, I am an associate professor in the ICT division in the College of Science and Engineering, Hamid bin Khalifa University. So I will try my best not to use my academic style of giving you a lot of technical things. I'll try to introduce at least what the cloud is about and how adoption is happening. Then I would like to take you through a story of a medium-sized company that has attempted a cloud first approach. And after that, I might go into a little bit of the technicality of virtualization and what it means. And then I will move into exploring with you and introducing this notion of edge computing. And I would like to end with a personal perspective maybe from a small business perspective. And I will be happy to receive any questions at the end before we move to Al-Hasan. And I'll try my best to be on time, inshallah. So what is the cloud? So the cloud, if we want to really talk about it, it's a model for enabling ubiquitous, convenient, on-demand access to resources. So you are able to access resources such as network, storage, computation resources. And you can access it over the net and you can provision it based on paper use. So it's cost effective. You don't have to have the initial payment, as Yazan said, it's elastic. So you can grow your footprint as you get more customers and you get more traction. And it's paper use. So it's a different business model to what we are used to in our standard hosted services. And some of this was triggered really by the massive scale applications, applications that are really generating thousands, million, billions of transaction each day. And it was also triggered by the ease of programming some of these cloud infrastructures. We have a lot of open source projects that are coming out to program, MapReduce, Hadoop and all of the new data storage systems, no SQL. And there's also the most important aspect that really drove this, which is the data intensive nature of a lot of application. A lot of our applications now are generating terabytes, petabytes of data, of daily logs, forensic, web data. And for us to process all of this data takes a lot of resources. And this data is something that you can monetize as a small and medium business. It can help you personalize. It can help you process ambitious thing. And target your customers in the best way. So in terms of adoption, in 2010 it was projected that the cloud computing market will move from 40 billion to around 240 billions as of 2020. When I checked at the end of 2020 this number actually was met and even exceeded around that time. And now between 2013 and 2018 there is a huge annual year-on-year growth of 30%. And we already have seen this in the last five years and the pace has accelerated due to the pandemic. So a lot of companies now are moving and governments are also moving to the cloud as a platform. So many providers that are available, so many open source projects, so many services that are being offered. In terms of the local market, according to the Ministry of Transport and Communication in 2018 the market was estimated to be around 33 million. And this is broken into software as a service, 53%, 42% infrastructure as a service and 5% platform as a service. So we'll talk about these in a minute. And in 2020 too it was expected that this will be growing to reach 112 million, 112 million. There's huge awareness, 76%, but the adoption was small in 2018. But it is accelerating now, especially with the Qatari government having a deal with a cloud provider, Microsoft Azure, to have a data center in Qatar and have that comfort of mind and providing this as a service to the government and probably to the private sector at the same time. So without further ado, I would like to talk about a case study. This is a case study of a new semi-government entity. This is a case study that we discussed in our cloud computing course in Ahmad Muqalifa University. And it was actually not developed by myself. It was developed by CIO, IT director of that entity who started with the entity. This entity is a semi-government entity. He calls it a startup. So this is all his sort of ideas there. It's a forward-looking organization, organization that wants to adapt the latest technology, medium-sized organization. They have a greenfield implementation. They don't have any legacy systems. Users are mobile, agile, just because the nature of their job requires them to travel and move around. And they are scattered all over the world. It has a headquarter, but definitely it's not a formal thing in the sense that they have a lot of multi-temporary and remote offices where they conduct business. They have no time critical application in the sense that most of their application are the general office use applications, things that are really not performance of time critical. They had options. They had an option to invest in their own on-premise IT infrastructure where they control the data storage systems. They have their own internal network. They set up their own security with firewalls and to have their own thing. And this was for them bad in the sense because they have to invest in hardware upfront without really understanding the needs. They have to handle all the security, which is very challenging and time-consuming task. There's a lot of do-it-yourself. They have to really control everything, the heat, the room, the servers, and they have to really employ a lot of people and invest in this on-premise infrastructure. But, however, the full control that this would have given them, that was a plus for them. And they were really thinking, if we had security attack or any issue of that nature, we can just pull the plug. So there is pros and cons for this type of on-premise infrastructure. But they didn't choose to go this way. They chose to go for a cloud-only or a cloud-first approach. And the promise for them, and this is using their own words, the promise for them at that point that they will get access to the best hardware virtualized, but they will have access to the best hardware. The systems that they will have as software will always be up-to-date, will always have the batches. As an entity, they don't have to worry about that, which is one of the main reasons to be compromised. And it also helps them to test without huge costs. They don't have to buy and buy and provision and hire. They can just start using it and with small staff they can start testing things, depending that on the other side, somebody is taking care of things. Security also is unparalleled. A lot of people think cloud is not secure, but it's almost the opposite where cloud infrastructures deploy the best security measures and they have the best security staff. It's centralized, well-paid, very high-end security. Also the promise of the cloud, if we go back to talking about the different modes, if they are doing it on-premise, they have to really control the whole thing. They have to provide the hardware, servers, storage, network, infrastructure and software. They have to provide the virtualization or operating system, runtime, everything. So all is do it yourself. While if they move and when we say others, here we mean the cloud infrastructure, they can depend that the cloud will, in the infrastructure as a service, will handle the hardware and the virtualizations for them, while they can maybe innovate in the middleware, running on top of it and in their application side. If they go for a platform as a service, they can also delegate more to the cloud provider to give them the basic middleware and the runtime. And then they can focus on the application and the scaling will happen automatically. And finally, they can use the software as a service, which is a lot of the services we got Microsoft 365 and Google Docs or whatever. They can get everything to be running and application on the cloud and to be using it as software as a service. And also moving from the on-premise to the infrastructure, they also have to deal with the payment instead of investing priori and really having the infrastructure on-premise putting a lot of capital investment. They are now having to move to think of it as rental, to think of it as paying for the CPU or the storage systems few cents per GP or few cents per CPU cycle per hour. And this is a new way to think about the payment and to think about utilizing this. And in a more technical discussion, we can really quantify it. If you really know how much demand you have, you can say what would be the best way to move. Is it on-premise infrastructure? And a lot of small and medium businesses usually do not have deep and clear requirements. So the cloud approach makes a lot of sense. A lot of providers are there, Microsoft, Google, Amazon and also Asian providers like Alibaba are coming to the market. But the main issue that they were worried about as a semi-government entity was the data security and privacy. Also because really they wanted to control their data. They didn't want anybody to have access to these jewels that they are creating and building as their business intelligence. They also were worried about residency where the data now resides in Qatar or outside Qatar. And this issue was important for them. They also, if you are a company that has investment in hardware, this is also an important issue. The third issue is really the skill set to be working in the cloud are a little bit different and this might not be as readily available for some companies, especially people who are trained on the normal IT infrastructure. So their approach, the company's approach, was to address these challenges, look at very extreme scenario, important scenario, for example, unauthorized data access from the vendor for some reason or the other or inaccessibility to the cloud infrastructure in case of natural disaster, no internet, something of this sort. And they will really think about these very extreme scenarios and build the systems to deal with it. And then legally ensuring that they have the relationship and the legal agreement with the partners. And they had big long-term agreements with cloud providers at the level of the organization, also at the level of the government. The Qatari government is investing with Microsoft now to move the government sector, at least to use it as a data center and it will be local going forward. They also had this idea of true mobility. The company is mobile. People are moving, no real attachment to physical space. And they have to really provide all their services with this in mind. This is common and natural for us now after the pandemic, but at that point, four or five years ago, they were really thinking about all this thing. So what they had, they built their system where they have a cloud provider. They are using infrastructure as a service for their storage and for their compute. They also use software as a service for some of the Microsoft 365 and they went with Azure as their platform. And really instead of having everything in on-premise, they have very light infrastructure on-premise, a few light servers and a network to do the printing while most of their asset lives in the cloud and there is a synchronization between the data that lives in the cloud and the data on-premise because they always wanted their data to be available to them if anything happens. The data was also encrypted with their own encryption key just to make sure that nobody can access their data and really this enabled them to have this mobility first approach and to move with Azure as their platform as of choice for a cloud system. This was the case study. I hope we articulated the point of this company moving to a cloud first approach, having most of their assets powered by the infrastructure and having a very light system on-premise to ensure the printing and the basic necessities are there to deploy things successfully. What's next for them? They are using now the cloud but they understand that the cloud has much more depth into it. They if the application demands increase they can utilize the cloud scaling up and down of resources. They can use and customize their code to optimize with the cloud network and the cloud storage systems. They are also thinking about multi-cloud approach where they can pick multiple providers or providers in different geographical locations and this will allow them to have redundancy and to handle failures and to do this they have to really even write the code differently. They have to use containers in order to deploy in a simple way and really this will require some more thinking and tweaking of their code. Also in terms of applications that are related to the location surveillance systems, sensory systems because they are managing a space where they are actually having a lot of these what we call smart city systems. For that they are also thinking about edge computing as something that can accelerate the speed and that can fit the need for that direction more. We'll talk about these ideas in the next few parts of the presentation. So let's move on and talk a little bit about technicality. Here we are talking about VMs and containers and the idea, just the idea of virtualization because that's one of the most fundamental idea on which cloud computing is built. If you look at a typical cloud environment you will have users accessing entities, accessing the cloud through the internet, having a security system and an interface to access the cloud and then having the cloud as a pool of these resources, a pool of servers, storage systems, cloud management systems that will allow the user to access this as similar to how we access electricity now to sort of connecting to this grid. So internet connectivity requires them to have a lot of security in place and then the load balancer can spread the traffic across multiple applications. And then let's look at virtualization. The virtualization is a technique to really put a mask on top of this physical infrastructure. So instead of thinking of a CPU, storage, CPU and specific architecture, hardware and software, you are thinking about these virtual machines that provide you access to this virtual hardware. So you can interact with the resources in a more straightforward way. So you can make one single physical resource appear to function as multiple logical resources or the other way around, making multiple physical resources appear as a single logical resource. So what does that mean? We have one physical server and on top of it we can deploy virtual machines. And these virtual machines, we can deploy multiple of them on top of the same physical hardware and this will help us reduce the cost by utilizing these physical resources better, consolidating these services into the fewest number of physical machines. We can also think about it from a load perspective. Different virtual machines are running different applications. They have different resource requirements as we show here on the left. And when we condense them, we put multiple of these virtual machines on top of the same physical hardware. We do what's called multiplexing. So the peaks and the valleys and the load are sort of covering each other. And the overall need does not add up to the whole initial needs. So by aggregating multiple workloads, you can really benefit and have a higher utilization of this cloud infrastructure and really allow the cloud provider to sell these resources cheaper to the companies. And this is really the benefit of multiplexing that adding A plus B does not add to the peaks that you would expect, which is the benefit of multiplexing. As a starting point, before we do this virtual machines, we have physical hardware, x86, process, memory, chipset, IO devices. And these resources are usually underutilized in most IT infrastructures. It's really in the low tens, 10, 20, very low utilization. And you have the operating system tightly coupled to the hardware. You have one operating system running on the infrastructure. And the operating system sort of controls the hardware. If we go to a virtual machine, it's the software abstraction. It gives you the interface of a hardware and it encapsulates all the operating system and the application state running on top of it. So it introduces this level of indirection, decouples hardware from the operating system, isolates, enforces isolation between the different virtual machines and multiplex this hardware across VMs. So it gives us essentially isolation where each virtual machine has its own domain. So the performance is isolated across these virtual machines and also issues related to bugs and everything. Security things are also isolated. It encapsulates the state of the application and the operating system. So you can easily snapshot and clone and move these virtual machines around. It's easy and portable to move things around because of this encapsulation. You can also do changes on the fly, transform some of these operations for memory and IO in order to really control what is happening and do some extra services like replication, compression, below the virtual machine and operating system level. So this virtualization allows you to have very nice properties that allow you to move things around and intersect and add extra services on the fly. We have system virtualization, which is usually called VMs, virtual machines. And in this case, the hypervisor, which is the virtual machine, run on top of the infrastructure in type 1. This gives you very good performance and there's a lot of virtual machines that use this. So you have infrastructure, the hypervisor on top of the infrastructure and then the virtual machine instances running on top of these hypervisor on guest operating system. Type 2 is what a lot of us do when we do it and we work on it. We install a virtual machine on top of our operating system. So you have infrastructure, operating system, the host, and then a hypervisor. And on top of that hypervisor, you have multiple things, usually in cloud providers. They allow you to do both, but the native bare metal approach is the one that have better performance. There's also the idea of containers, where we are not virtualizing the hardware, where we are actually doing operating system virtualization. These containers or zones, whatever you call them, they run on top of the operating system. Some of the goals are satisfied. You can segregate the performance of the apps, easy start, stop, move and management of the instances. And you can do that without the full virtualization that I told you about. You can create these zones or containers. They run on top of the operating system and they allow you access to the resources and the resources within these zones are segregated. So each zone has its own applications, its own networking stack, its own addresses, ports and user accounts. So you can get the benefit of isolation or semi-isolation. CPU and memory also is divided between zones and containers. And so you can have your own scheduler. So you can get some of the benefit of virtual machines, but you are doing it on top of the operating system. And if we want to compare this, we can look at the ultimate separation in the bare metal to run different virtual machines on different hardware. Then you get virtual machines where they run on top of the hardware and they virtualize this hardware, giving you multiple virtual machines on top of the same hardware. Then we move into containers where you can have multiple containers running on top of the operating system and giving less isolation, but relatively robust system. Or you move toward the end where you can allow all your applications to run over the same operating system without even the boundaries that container give you. So to the left, you have less management burden, everything is trusted. To the right, you have better isolation and this is very important for this mode of operation where you are sharing your hardware with applications that you don't necessarily trust because you are running in the public cloud. So in the public cloud, a lot of solutions use virtualization, hardware virtualization, virtual machines, and on top of virtual machines, they use containers to maybe give more finer control to different applications. Containers also is something, and I wanted to mention this because a lot of us might use containers just because of the software perspective. They give you a very nice standalone unit of a software packaging. You can put all of your code, all of the libraries dependency in the container so you can run quickly and reliably in different platforms. So it's a packaged software using standardized units of development, shipment, and deployment. And this container is really a software engineering sort of shipment and deployment idea. It's really different from the idea that we presented initially of containers being an abstraction that virtualizes the operating system and gives you this isolation, even though they are related and connected but they are separate. So containers gives you both benefit and this software perspective benefit is the main reason a lot of us are using containers. So I hope at the end here I introduced you the cloud environment, virtualization, the differences between hardware virtualization, software virtualization, containers, and Docker is the main one that is known for these container platform. And I hope most of you will be considering using cloud application, cloud infrastructure. If you are interested, you need to pick your provider, Amazon, Microsoft, Google, familiarize yourself with the pricing, models, and tools to manage virtual machines and to access the hardware storage and memory systems. Utilize the cloud. Initially, it's much easier just to use the cloud as if you are using hosted IT infrastructure. And then as your application develops, as you understand the cloud more and you adopt the cloud mentality, you can look at how you can restructure your application to fit the cloud infrastructure, the networking and the storage infrastructure. You can also look at services in the cloud, a lot of clouds give you machine learning cores, gives you quantum, gives you specific service for media and different other applications. So I hope this introduced you to the minimum that we can talk about when we are thinking about cloud. I also wanted to talk to you very briefly about edge computing and about our experiences with studying this new approach of managing IoT and smart city systems. Edge computing, if you think about the cloud perspective, it's a process where the cloud offloads some of its compute storage from the centralized cloud, usually at the cloud data center level, to the network logical edges. Usually the Internet service providers control or the communication providers, ISP providers are where the edges are deployed. So you deploy these edge server closer to the end user. So different applications, mobile, smart cars, sensory applications can access. And the edge computing encompasses a lot of technologies that were coined, like mobile edge computing, fog computing, cloudlets. And essentially all what it means is that moving some of the resources, computing, and network resources along the path between the data sources and the cloud data center. So your applications become hierarchical where you have different layers doing different things. And we'll talk very briefly about what these layers might be doing. Potential for these applications, you have smart city, intelligent transport, smart grid, smart health systems. There's a lot of applications that can benefit. And the application will benefit from better latency, reduced traffic, better utilization of different resources. That's the potential. And there's a lot of challenges. There's security privacy challenges of moving this computation closer, resource management, and a lot of other challenges that are addressed in research and in the cutting edge providers, which are usually now the ISPs. We talk about our experiences. We have studied video streaming as a potential application can benefit from edge computing. Here we look at live and on-demand streaming of video. It's becoming very common on the internet. For example, Facebook Live has 2.9 billion active users, has 78 percent of Facebook users are watching live video, one out of five videos today or around that study in 2018 was live video. We studied with funding from the Qatar National Research Fund. We studied scenarios where you have multiple people capturing an event, uploading this live video to the cloud infrastructure. And the cloud is a multi-cloud infrastructure that is distributed in different countries and locations. And users also are distributed in different locations and countries. And now, when you have this complex problem of multiple video capturing crowdsourcing and multiple viewers in different locations, you should consider where to deploy your transcoding, where to serve the video from, which is not an easy problem. And also here, we developed tools and techniques using standard mathematical optimization tools and using AI reinforcement learning to do the resource management and provisioning. We also deployed edge computing to deal with caching and processing. So now you can think of the towers, the telecom towers as edge computing. And then you can do some caching and some transcoding. And you can allow multiple of these towers to do collaborative caching. If the video or the chunk of the video is there, it can be, you can get it quickly. Or if it's not there, you have to go back to the cloud. If it's there with a different quality, you can transcode it and send it to the user. So you can do very interesting things in terms of collaborative caching and processing. You can also have a multi-layer architecture where you have a cloud, an edge, and then the end devices. And you can get all of these layers to really complement each other and provide the service in the best and nicest way. Another application that we looked at very briefly is the smart health systems. And in smart health system, there is huge growth in terms of the expenditure for chronic diseases. Globally, this is the figure in the U.S. as you see here. And Qatar also, total expenditure is 2.2 of the GDP for chronic disease. Death caused by chronic disease is 76% of death in Qatar. Percentage of people over 60, aging population is on the rise, which leads to a need for smart solutions. Traditional health care is really not scaling as much as we want. In any hospital now, globally and locally, you have lack of rooms. You have people who have been let go out of the service very quickly in order to have that bed or these resources dedicate to the next patient. Now there is a move globally to use sensors and mobile technologies to complement traditional health care. This is going to allow efficient, convenient, continuous monitoring that is remote. And the patient will participate in their own treatment, which is called the Internet of Medical Things with powering remote health monitoring. Edge computing will be very critical in some of these scenarios, where now you are collecting data in real time from a patient that is at home, doing some preprocessing. As you see here, the Edge network, you do some data acquisition, you reduce the data, extract the features. If you're doing some sort of machine learning, understand the context of the patient. And based on that context, you can decide whether you want to send more data, the raw data, or compress the data, or just send the features. So somebody in the cloud side will run the machine learning, data reconstruction, and give some sort of analysis and treatment, alerting the medical provider, the doctor, which can then see if this patient needs to come back to the hospital or it's safe to stay at home. So Edge computing here will save you delay, bandwidth, energy, because you will control how much data is sent, you will control a lot of things, you will send alerts, and you might have most of the data not being sent and keeping it secure at the patient side, if it's not that critical to be sent. There's a lot of application, I will not spend a lot of time here, but the same mentality of Edge computing can be applied to smart mobility systems, surveillance systems, emergency response, smart cars, UAVs, all of these can be connected to the Edge and these edges can be connected to the cloud and we can get the same benefits and get the same advantages. And here in Ahmed Makhaneef University, we have a lot of work to do these applications, smart city systems, AI, running on the Edge to do some very interesting services. And I wanted to end here with a personal perspective. In 2014, I started a small e-commerce company, the website, I don't know if you guys remember, Qatar Best Deals, and we deployed this. And when we started this journey, and this is the perspective of a small business, we were a small business, few people, five, six, started with an incubation and the digital incubation center. The core team was in Qatar, but we also had some development happening internationally in Tunisia. The company started, we started growing, sales started growing steadily. Initially, we used the standard web hosting service. That's what most smaller companies would use. But really, very quickly, we hit a wall where a lot of these hosting services give you a quota or very rigid type of thing. So there was no flexibility to grow and there was constant limitation and sometimes service failures where we exceeded our quota as we grew this company. At that point, we were incubated in the digital incubation center. We were introduced to Microsoft Azure. It was initially for free to be used and to experiment with when we were in the digital incubation center. And really, we felt very relieved because we were able to scale our services. There were no ad hoc limitations put in place. And this was our back end, our core e-commerce engine. Also, as a company that was distributed, at least one main location in Qatar, one main location internationally and also few people working remotely within Qatar. We used a lot of tools that now are becoming the norm. We used Asana for remote collaboration, project management. We used Dropbox, Google Drive for file sharing, Skype for video conferencing. WhatsApp was used heavily also for messaging and keeping track of the daily updates. And all of these are software as a service tools that are really utilizing cloud infrastructure in the back end. So also code repository bug reporting and cloud providers. So that happened before COVID. Now this is very common. A lot of companies are online with all of their work. So really, for a small business, this approach makes a lot of sense. It allows you to be more agile and to provide very nice complement. As a summary, I want to say that cloud computing is a key enabler for small and medium enterprises. Adoption is happening, and it's growing rapidly in the public sector and in the private sector globally. It already happened in Qatar. It's also ramping up. Companies need to develop skills and know-how in understanding virtualization and distributed computing, but not let that be an obstacle. Using clouds is fairly simple if you invest a little bit of effort. Edge computing is also enabling real-time processing closer to the end user, powering a lot of these smart city applications and futuristic applications. I would like to thank all of you. I have finished my material. If you have any questions, I would be happy to take them now or as the host sees. Thank you so much, Dr. Ayman, for this very beneficial session. We received a couple of questions that I would like to share with you. The first question was, what type of data that should be considered for clouds? Yes, let me elaborate a little bit. Some people are really worried about using cloud computing because some of their data might be confidential. Let's talk about multiple answers. The easy answer is you have to really make sure that whoever you are providing the service to, usually the government does not have reservation from you using a cloud provider in the sense that if the data center is local and you are allowed to access the data, then hopefully if you get the permission from whoever you're giving the service to, to utilize it, any data is possible. Another way to answer this question is some people build hybrid system where they will have part of the data living in their data center, very sensitive data, credit cards, maybe personally identifiable information, things that they don't want to put in the cloud. And part of the data lives in the cloud, the part that is fine, the part that is handling the core business and my benefit from the cloud. So that's another way to think about where you have to decide what data you want to keep local and what data you want to put on the cloud. The last part I want to add is some companies will like to use the cloud for all of their work, but they do not want the cloud provider or anybody who can break into that cloud if that happens to access. So they will encrypt all of their data but have a copy of it in the cloud. So I hope this multiple answers your question. Thank you so much, Dr. Ayman. One more question that we received is running Java applications over GPM, an example of quantization? Java virtual machines? Yes. Yes, it is. I only talked about two types of virtualization but when I teach my course, I talk about four types of virtualization. There is hardware virtualization, virtual machine. There is operating system virtualization, which is containers, dockers. There is also what's called a programming language where you are virtualizing. You are creating these bytecodes running on your Java virtual machine. And this is also a sort of virtualization because these Java virtual machines also have some sort of isolation but it's a very low level isolation. It's sort of in the low end, which is isolation by running separate processes on their own JVM, which protects you from some things but also there is no performance isolation because you are using the same CPU, you are using the same, there is no boxes that can isolate these two things from each other. But we consider it a type of virtualization. We call it programming language virtualization. Thank you so much for today. And what are the differences between the two entities? Yeah, I see the question of hypervisor and virtual machine. So I use them... I use them... If I'm teaching you an operating system course, maybe I would deal with it differently. But for all practical purpose, the virtual machine monitor is the hypervisor and vice versa. So they are the same for most people. Yes, thank you so much. And should companies not do data classification and risk assessment before moving to cloud? No, I think they should do, actually. They should do. That's why if they really have some data, where the entity they are providing services with will have some mandate, not to have that data in the cloud, that should be respected. Now, in Qatar, there is a lot of rules that are being deployed in the government sector of asking each entity to define what is private data that nobody should access, what is public data and what lives in the middle. And for different type of data, the rules are different. And even these are rules are mandated by the government to have some sort of a unified system across different government entities, mainly for security purposes. But this is becoming the norm now. We are going to go in the future in 10 years from now. All of us are going to use the cloud. The cloud is going to be everything. I am monitoring the IT business for the last 10 years. It was a huge business. Now the cloud is really consolidating all of this IT into these central systems that are managed centrally by big cloud providers. So people have to know where their data can live. They can have these hybrid solutions, but definitely there is no way backward into moving into the cloud going forward. Thank you, Dr. An-Mashallah. It's fruitful discussion. We have a lot of questions for your session and we should keep track of the time. Of course, the audience and they would like to contact you. I'm not sure if you would be able to please to share with us your email address for them. Yes, I had actually one slide here at the end. I would like to acknowledge our funder, Qatar National Research Fund, but also this is my email. I'm an academic so you can reach me easily. Just type Iman Arbaad my name and you will find my email in the university website. You can find it here. LinkedIn is also the best platform, but also other you will find so many online personas from where you can find it. Yes, so Dr. Amen, one more question before we move to the next section. Is there any government platform that small businesses can use to create their business online? I don't think there is a platform, but there are communities and there is support. So Digital Incubation Center, not because we are here with the MOTC today, but really that's the place where I when I started that company and I remember I've seen Al-Hassan at that point. That's where a lot of companies are being hosted, provided a lot of services, even some access to some limited cloud trials. There's a lot of support there. There's also support in the other incubation centers that are available locally, QSTP, but that community of DIC is very a strong community. I would recommend you to join and at least be affiliated with them. Thank you so much, Dr. Amen. We received one question that small businesses need help in accessing cloud source. I would like to answer that question. Our program that we are running this digital forum for and you can access our website that I will be sharing it in the end of our session today. The project, the DTSME, the Digital Transformation of SMEs, we run daily webinars on the importance of digitalization using cloud services and we provide and offer special rates for SMEs on cloud services and e-commerce and online presence. So I am inviting everyone who is attending our forum today to participate and engage with our daily webinars. We'll find more information on our website on motc.gov.com under the DTSME project, the Digital Transformation of SMEs. One last question, Dr. Amen, before we go to Mr. Al-Hassan. In 2018, we could see 3% conversion or interest in the movement to cloud. How is the right now? This is a tough question, but I can assure you due to COVID and due to people realizing how they can work, the having companies work purely online, this is really accelerating with a huge pace. And also in Tatar, particularly after the government last year had made it very clear that there is a long-term partnership with a cloud service provider and that data centers will be local here. The rate is really accelerating. This is good news for us because if the public sector starts investing, the prices are going to go down and the experience will become more widely available. So I do not know now, but I think it's going to accelerate strongly. Dr. Amen Erbid, Associate Professor from Ahmad Bin Khalifa University, thank you so much for your time. Thank you for this beneficial session and the great information and knowledge you have shared with us. Thank you so much. Thank you. We would like to move ahead now to our next session with Mr. Al-Hassan Samurra'i, the founding partner and general manager of APLAB. He will be taking us through the advancement in technology and impact on business growth. Mr. Al-Hassan. Yes, good morning everyone. So let me share my screen. So I don't know if you do see my screen. Yes, and of course Mr. Al-Hassan, in the end of this session, we will be accepting the questions on your session as well. Sure, sure. So good morning. This is Al-Hassan. I'm very glad to be here with you guys in MOTC and HVKU and of course our participants and audience who are SMEs. And I considered myself for some time to be an SME, so I feel that I'm one of you addressing you and I'll be glad also to interact with you. So I'll start with a small introduction to APLAB. So APLAB is a software development company. We claim to be the biggest or the largest software house that resides here in Qatar with more than 75 plus employees and professionals in software, in data engineering, data science, security and business analysis. So we started in 2016 with the value proposition of having a team in software that resides in Qatar with no outsourcing and no freelancing. That was not there in the market. And still we claim to be the largest software house that have these values in the market. We do business process automation, security, design, cloud, mobile and web platforms. And of course the main topic of this session here is data. And I'll try to start from where Dr. Aiman have stopped, where I will give real life examples of entities here in Qatar that is doing cloud and is doing data engineering and science as well. Now, in general, we are in an era where there is an industrial revolution and what Dr. Munir have mentioned that we are in the industrial revolution for that is mainly around smart systems and smart technologies. From all of these technologies, you name it IoT or cloud computing, data, business process automation, among others, cloud and data-driven intelligence are, in my view, are the most enabler technologies that will drive all other technologies and all of the advancement in industries, not only in the previous years and the current years, but also for years to come. We'll start with some explanation that will get us into the topics which we want to and the examples which I want to elaborate here. So cloud computing, I think all of us knows what cloud computing and Dr. Aiman have got us through some good literature. But a lot of us thinks about cloud computing as outsourcing the data center. So instead of having a data center, let me use the data center of someone else or another company. So yes, it is, but currently cloud computing as a terminology is changing in terms of its meaning. So when we say using the cloud, initially, this term have implied that we'll be using a data center of a company that is well-managed. So we'll be using virtual machines there or storage there. But currently, when we say we'll be using cloud computing, it means we will use the services over the cloud. What this means is that if I need to have a platform, for instance, I'll not have a virtual machine over which I'll build the platform. It means that I'll directly go and check for that platform as a service over the cloud. And this implies not only to platforms, but also to softwares, to all of the components that any software or any technology requires. And we'll come through examples that will show you that this is what currently experts and entity means when they say we want to take this cloud. It means we want to use the capabilities of the cloud to implement solutions. Now, when it comes to SMEs, of course, the first meaning that is the initial cloud perspective of outsourcing the data center is very important. But also having everything and almost everything nowadays as a service, as a software, or as a platform, is really what can enable SME to focus on their business. So instead of focusing on how we will do this, how technology will enable us to do it, you will be focusing on business and leave the rest to the cloud. And this takes us through the infrastructure as a service, platform as a service, software as a service, explanation that Dr. Aiman have touched upon as well, that you will not be looking for virtual machines over the cloud to enable you as an SME to put your platform there or to put your software or to architect your software. You will be looking directly on what you need. And this varies in terms of service. So some of you will go to platform as a service as they know how to deal with a specific platform that they want to be readily available. Examples to this will be Oracle Cloud, SAP, Hadoop. But some of you will go directly into a software. And the software service means that it's something like when you directly open a Google Sheet and start to work over it, or open Office 365 Sheets and start to work over it. That's the direct software service that doesn't even require a platform to be managed. What I'll be doing here is I'll be pitching different architectures where people used to struggle to implement this on a data center or even an outsourced data center. Examples in Qatar are Uridu data center, for instance, or Mize data center. These are data centers that resides here in Qatar locally. They provide you the data center functionality where you can say that I want this storage. I want these virtual machines. I want three virtual machines, two different sets, one for production, one for development. But then also you need to do everything around this. Now, what Quarante Cloud and all of what it means provides is services. Here is an example of what do you need to properly manage software. So when you have software components over your architecture, you will need to do proper version control. You will need to push these softwares to a production environment, do the proper testing, and then push it to the production environment. You might have different engineers working on the same piece of code. You might be using containers, for instance, that Dr. Aiman also have just mentioned. So containers is a way to contain your codes, not to be affected by the hardware level or the virtual machine's level. And it's also a way to enable you to scale and do high availability setup easily and cheap as well. So managing containers, managing version control of codes over get repositories, pushing all of this to development environments and production environments. All of this, if you wish to set it up properly over a local data center or an outsourced data center, it will take days and weeks. And for big enterprises, this used to be multi-million dollars industry. Now over the cloud and using services, you can do this neat setup in a few hours only. And that is the beauty of cloud. And that's what cloud should bring to your attention when you are managing a startup or managing an SME, or even managing a government entity similar to the use case which Dr. Aiman have just presented to us. This is another example of an architecture that is a typical architecture which we in App Lab are delivering to a number of entities in Qatar. Now if you see this architecture, it starts with a web service for the backend software basically. And when we say web app service is basically a platform as a service over which you can put your container that have the codes and immediately start to run your platform. Now the platforms in general doesn't work in silos or isolations. You have to have so many integration for a platform to work. So your software might need to send SMS messages. You will need to transact payments, for instance. You need to serve through web services, mobile applications over iOS and Android. You might need to have single sign-on. So you might need to integrate with an active directory for your employees to have single sign-on. And you might want to integrate with other systems that you have internally, whether it's an ERP or so, through a middleware or direct integrations. Now to do this over the cloud and maybe this architecture can show you clear examples where Azure Active Directory, for instance, is the Microsoft Azure Cloud Active Directory that used to be domain controllers over data centers. Integrating it, integrating an app service with Azure Active Directory to achieve single sign-on is kind of an implementation that is already available in the cloud. And for those of you who know who have been there for a while in the industry, they know the pain of having integrations with systems. This is something that is currently over the cloud. I will not say that it doesn't require expertise or so, but it is an easier job that enables SMEs to achieve their goals easier and cheaper as well in a lean approach where SMEs and startups demand for. Now, although it can be as simple as this, it can grow as complicated as this. And this is for your information is the Azure Typical Architecture, which is the cloud typical architecture for government currently. We'll not get through this, but although this seems to be very complicated, almost most of the boxes that is shown over this architecture is a service. It's not a virtual machine or an infrastructure that you will need to manage to deliver your solution or have it up and running. This is another example of a system that is architected over the cloud. So another container approach other than the Docker container, which has been mentioned by Dr. Eamon, is Kubernetes. And for those of you, for example, who knows Pokemon Go as an example that have been launched a couple of years ago in the initial days when Pokemon Go have gone viral, you have had to wait. So you log in over Pokemon Go and then you keep trying to log in for some time until you get your queue and then you start to use it. And the reason was that although they built this over Kubernetes containers over the cloud, it hasn't been configured properly. And that expertise was very much a major at those days, which is currently have developed so much progressively with experiences such as possible Pokemon Go. And this is by the way an example of an architecture which we are delivering to an entity here in Qatar. That is a government entity where we are building a software to be scalable over a Kubernetes container with a Kubernetes orchestrator. And there are so many data engineering and data science components which I'll come to so explaining how data science and data engineering is very much directly related to the cloud. So basically, if you want to take a takeout out of this is clouds enables for high availability and scalability. And normally in the industry, high availability and scalability used to be terminologies that contributes to high price technology, which is currently not the case using cloud services. Now, when it comes to data and how data is related, data science used to be a terminology or a technology that have been there from conceptual and theoretical perspective for some time. Having this said, it has been only enabled recently when computational power becomes available, when we can have millions and millions of computational operations within milliseconds. Although computational power was the enabling factor, cloud I would say is the second enabling factor for data science and data engineering. Today, you as an SME having a small e-commerce system still can have data pipelines over the cloud, have proper data analytics and visualization of our solutions like Power BI, for instance, and have proper data science machine learning algorithms that works over your data without complicated deep knowledge into data science. And that is the beauty of it. Now, when it comes to data, I'll just try to start with some basic concepts. So you are running an SME or a startup, you have to ask a number of questions before articulating what data and what data engineering components and data science components you will need to do data management over the cloud. So first you define your questions, like what do you want to achieve? Do you want to achieve more transactions over your platform? Do you want to have analytics only and visualizations which will enable you to take decisions in your business? So you define your questions, you start collecting data. So you try to find where is this data? Is it over your, let's say, margin to a framework which you are using for your e-commerce platform? Is it over Google Firebase which you are using to manage notifications and real-time messaging, for instance? And you start to find the collectors over the cloud that can enable you to aggregate this data in a data link. Now, throughout this process, you will need to do cleansing and processing to really have a meaningful data that you can use. Now, although you can go all the way and do very complicated data science modeling over this data, but initially simply you analyze this data and visualize it over any tool that allows you to visualize it, one of which is Power BI, for instance, over the cloud. Now, we have been talking about so many data components and we have gone through some real-life cloud implementations and you have been hearing data engineering and data science. And I'll need to explain this in simple ways. So data engineering is basically the pipelining methodologies. So how you will be able to aggregate data in what form of database or data warehouse from where you will be able to extract the data, transform it, store it. All of this is data engineering. Now, data science comes when you need more sophisticated processes, processes where you need to get meaningful information out of your data. You want to know, for instance, that people who are of certain group, per se, have in general tendency to do orders at certain hour, for example, within the day, or have tendency to like certain cuisine. This is the type of questions where you need to apply data science and machine learning methodologies over your data to get this meaningful information, which by the way, again, you will need to engineer it in a way to save this data and then present it as well. Now, there are so many processes that are related to that and it can be very much complicated. It can be very much easy as well and straightforward. Having this said, there is a statement that always you can use to to explain the differences between the layers that are actually engineering and the layers that is data science. And that is a very well-known statement by David Bianco, which says that data engineering are the plumper. So the plumper who built the pipelines of data and then data scientists are the painters or the storytellers that can get the information out of the data. So what does this mean over cloud? Now, when we come to data engineering, the known terminology for this is ETL, which is basically export, transform and load. And every one of this is a process where if you want to build it over a data center, you need to set up proper software stack that can extract data from its sources like databases, for instance, and then transform. So you need data transformation layer, which transforms data into meaningful data that can be directly loaded, which is the L of the ETL, into analytics screens, visualization screens, or further data transformation layers. Now to do this over the cloud, it is again, as I have explained in previous architectures, it is as simple as using services. So for instance here, for data transformation in the ingress stage where you actually export data from its sources, you simply use an Azure data factory that is a pipelining and transformation service that is available over Microsoft Azure. Now, I apologize for using, as an example, Azure components. We are not marketing here for Microsoft, but it's actually, it's a country decision. It's a country decision to go with Microsoft when it comes to government and when it comes to building a data center in Qatar. And I'll come to this at a later slide where you will see why you will want to choose Microsoft Azure for your cloud. Now, apart from the transformation layer, there are a number of data warehousing and data leaks layer where you architect your data within. Now, the difference is that data leaks normally are used to store data in different formats. So you might have structured data, you might have no SQL data or not structured data like MongoDB, for instance, any real-time database. All of CSV files, so you might have CSV files as well. All of this can be dumped over a data lake after which you start another transformation layer where data can be transferred to a meaningful data that can be stored in a data warehouse that is actually an optimized layer to load data into dashboards and reports. So this is a typical cloud architecture of a data pipelining, data warehousing, and data visualization challenge or problem that an SME might face, a startup might face and of course, higher-scale enterprises and government entities. I think we have gone through the ETL stages. I'll try to get into more examples because I think examples can show you directly the meaning and the enabling factor of clouds when it comes to data engineering. Now, in general, of course, there are a lot of challenges related to data engineering and the way you will be dealing with data as an SME, one of which is quality. You might have a lot of data that cannot have meanings and doesn't have the real quality because of software decisions or technology decisions that have been taken initially. So you might have one or two years of data related to a fleet management system where you wasn't, for instance, storing information related to trips. And this is very much realistic. We faced this with a very big name, an enterprise that is very big here who was doing fleet management for the last three years over a very good software but which has not been meant to store meaningful data in the database. And this has been a huge challenge because when we came to do some data engineering and data science work to enable them to get more transactions, for instance, it has been very much challenging because there are crucial information that is not being stored. There is the challenge of integrating multiple data system. You might have different software components. You might be using different services currently. You might be, for example, doing your customer service over a WhatsApp group. And still there are information there that can contribute to your decisions. So integrating with the different data sources is always a challenge. And the metadata, of course, is always a challenge. If your data is properly stored but you have so much data and this data is scattered over different data types, it is always a challenge to get all of this into perspective and position it properly over data warehouse in a meaningful way that can be used in further transformation and loading also to the visualized layer or the analytics layer. I wanted to get into this slide directly because this is very much a typical architecture which is very much important for SMEs as well as other professionals in the industry to understand of basically the typical way of implementing a data engineering and data science project for anything that can be an in-house problem within an SME or an enterprise or a government. So here you can see that on the left there are the data sources and these data sources are of different nature and different structures coming from different sources. And then your main aim initially is to get all of this over a data lake. So get all of this over a lake of data that you can say, yes, I have now all the data and they are not stored over different repositories. Once you do this and you have your data lake, then you start to think about the transformations that you need to do to get more meaningful data into a structured database. And here we use the data warehouse because in general the data warehouse is a database that is scalable and optimized to be used for further transformation or for visualization over analytics layers. Now here again we are using a Microsoft Azure Cloud example. So the basic element for transformation over Microsoft Azure is Azure Data Factory. But still you might have transformation that requires more sophisticated processes for which you will also have readily available services over the cloud. And here we are using Databricks and HD Insights. One more element which I didn't touch here in this architecture is this. So for a lot of examples, entities, and it might be yourself as well in your SME, you might have data that are still not digitalized. And for these, you can simply use Power Automates and Power Apps as an example of data inputs, software front-end, and data flow management for approvals and review, for instance. Take the example of a form that you want to fill continuously and then have someone to review it and someone to approve it before getting into a proper database or data structure, for instance. This architecture pretty much covers all of the life cycle and journey that you need to have when you come through such requirements or such projects. Again, here are the main components that formulate such architecture over any cloud. And here we have taken the example of Microsoft Azure. This is an example of, and I'm looking here on the clock as well. I want to have some time for question and answer and to wrap these sessions as well. But this is an example of a Power BI, for instance, dashboard that is presenting information after all of the processing and transformation stages. And this information directly resides over the data warehouse. Again, although we mentioned some challenges but there are some challenges that directly related to SMEs. One is data migration again. So data migration, especially in the case where some data components resides over data centers, for instance, and lack of control capability. Now, having all of this said when it comes to the easiness of using services over the cloud, still the expertise is required. So someone doesn't need really to jump directly into implementation. You have to get through some courses and you have to have the software background as well to understand how to architect initially and then use your architectures. Now, in general, and I'm mentioning this because this has been a strategic move for Qatar where Qatar have decided to have Microsoft as a data center in Qatar. And what this brings to the table for Qatar is so many things. One of which is a challenge that has been mentioned by Dr. Aiman when it comes to compliance. And I think some of your questions as well. Like you need to do data classification and risk study. And then over and above this, you have your national information security compliance. And for instance, in the health sector, it is not allowed to store health information over the cloud or over repositories that doesn't reside in Qatar. One more example is QIDs. So your business that have a technology component might need to store Qatar identification number and by law over the codes of NIS or NIA. This is that is basically the national information security code out of Q-SERT. This is not allowed. So to do this properly using cloud, we have to have a data center in Qatar. And that was the strategic move by having Microsoft Azure. So you don't only have a cloud that resides in Qatar with all of its cloud benefits, but you also don't have problems when it comes to data that have to be administered under Qatari law within a data center that resides in Qatar. And that was basically what have been signed late in 2019 by Prime Minister at that time Abdullah bin Nasser. I'm quite done with my presentation. So again, there are a number of other things that have to be considered our core terminologies that we have gone through today was cloud and data. But also there are so many things that are related, which is business process automation, security, UI design, building the actual custom components, whether it's web or mobile, that is very much part of an ecosystem that works along with the cloud and that. I think I'll not go through details in this. So I think we have gone through enough of examples where we mentioned business process automation, cloud. And all of what we have just said is always related to front end day-to-day applications where you need to collect some information over a mobile application or you are doing transactions and payments over a web application, for instance. So thank you very much. I think I'll be happy to hear any questions you have, any comments, any topic of discussion that you want to discuss. And here are my contacts if you wish to approach if you want to have any discussion after this forum. Thank you very much. Thank you so much, Mr. Alhazam. Thank you for all this valuable information. We received a couple of questions that I would like to share with you. Is data engineering seen as data cleaning, processing and standardization? Yes. So data cleaning is a data engineering. Is a data engineering, basically data engineering is the wider terminology. So in data engineering, you set up your data pipelines, you do cleaning, you do simple transformation that are not actually machine learning or further surfacing modeling. That is the actual data science part. So yes, data cleaning is part of data engineering. Thank you so much. I received another question. So how can SMEs know or... Yeah, and you know what are the pros and cons of using private clouds offered from a technology provider to another? Sorry, Ravan. I received a question. And how can SMEs, how can they choose which private cloud service providers should they approach? How can they know the pros and the content? What should they consider before they're going to this particular service provider to purchase their cloud services? Sure. So basically, initial advice, so advice one, E or a startup, don't do to a data center provider. Go to alternatives and options when it comes to services that you will be using. When I say cloud, that is a cloud. I mean Microsoft Azure, AWS, Google Cloud, Alibaba Cloud, for instance. And then over these, you have to ask yourself the right questions. Initially, you might need some free credit and you might get it through some packages from AWS or from Microsoft Azure. But then you ask your question of what services do you need over the cloud? And when it comes to the services, you will need to look into currently what do you need and in the future, what do you need? And then look into these clouds and see what software components are available there and what you are happy with and familiar with doing over this cloud. Now, last point is if you have by any means crucial data and data that is sensitive, when I say sensitive that low might mandate that it has to be in Qatar, you will need to look into two approaches. One is to go with a hybrid approach that where you save this specific data on a data center in Qatar or over a component that resides in Qatar or go to a holistic approach where the whole cloud have a leg in Qatar that meets the compliance you will need at a later stage. Thank you so much. There is one more question. Yes, how is the spending capacity to adapt into cloud? So how should they consider their investments and their spending capacity to adapt into cloud? Right. Now, in general, if I understand the question correctly, in general pricing between clouds nowadays is kind of the same. So clouds are trying to match each other so that the competition is more of the services rather than pricing. Now, when it comes to your own pricing, so how you can really have an optimized budget because you are on budget, for instance, but you still want to have a technology setup. Definitely going to the cloud is the approach that will save you money because you don't have to have a big investment initially. But also you have to be wise in your decisions over the cloud. And I will give one example. That is the security layer. So you can always have the security layer that is a security layer, let's say, over Microsoft Azure or over AWS that matches government recommendations where you will need to have WAF, you will need to have proper application gateway, you will have denial of service, denial of service attacks. All of this to build it properly with state-of-the-art technologies that is available over Microsoft Azure, for instance, might cost you around $10,000 of resources only, monthly, for this security zone. Rather for an SME, for instance, you can go with a simple service from Cloudflare, for instance, which is another cloud. And nowadays also they are working on a service that is available over Microsoft Azure or AWS. So instead of paying the $10,000 a month, you will be paying $200 a month with reasonable security for some time until your requirements really elevates to a model where you really need to elevate your security levels and so on. So this is one example of a type of decision you will be taking initially to optimize your budget over the cloud. I think, Ravan, you are on mute. I don't hear you. Yes, thank you so much, Mr. Alhassan. We received a couple of questions how to use and integrate some cloud services and a question on what is our role in supporting SMEs and obtaining cloud services. Let me just repeat our digital transformation of SMEs program that is running. This digital forum runs daily webinars supporting and educating SMEs on the importance of the digital transformation where we are educating the SMEs and providing the SMEs with customized competitive prices through our service providers that you are working with, with the biggest companies in Qatar and the best technology providers worldwide. So I would encourage everyone and I'm inviting everyone to attend our daily webinars so we are covering the challenges the challenges, the opportunities, the support that the government is providing in collaboration with the Ministry of Transport in collaboration with the Ministry of Commerce with the Qatar Development Bank with our service providers who are offering services and like I said, competitive prices for cloud services, for web presence and for e-commerce. So once again, I would like to invite everyone to attend our webinars. You can find more information on our daily webinars on motc.gov.qa on the DC SME program, the digital transformation of SMEs. One last question, Mr. Ahsan, that I would like to unfortunately end our session together with. Can small organization adopt artificial intelligence and data science to increase the sales and testify cost and implementation? Yeah, that's a tricky question. Just before this, Raman, in the contents of the services you are providing, I strongly also recommend all of the audience to use the ecosystem MOTC is providing. We have been graduates of the Digital Incubation Center. We have been using a number of packages that have been provided because MOTC have very good agreements with cloud providers as well as other technology companies to give support to startups and SMEs as well. I also strongly recommend checking always this is another opportunity also to market for my university. HBKU also for the programs they have when it comes to deeper knowledge into data science and data engineering. Now, maybe Raman remind me of the question again. Yes, of course. And thank you so much, Mr. Ahsan for this edition and we really appreciate our partnership with Hamid Ben Khadija University and with the app lab. Can small organization adopt artificial intelligence and data science to increase sales and justify the cost of this implementation? Yeah, the answer and simple is definitely yes. Now, although always I am sensitive when small SMEs and startups talk about AI because sometimes I feel it's very early. So from the context of the question it's very clear that you want to enhance or to have better transactions to have more transactions over your platform. There are a lot of ways in the roadmap until you reach to AI. So simple data analytics, simple data engineering and some data science techniques that is not necessarily AI in the initial stages will get a huge ramp in transactions. And we have seen this. We have worked with a lot of names here in Qatar. Very big names I don't want to mention but they are local competitors of Talabat and we have achieved huge ramp in transactions simply by using simple data analytics. Now, having the same, yes, you can have artificial intelligence on that science and using the cloud services even a simple startup or an SME can have an AI engine that is running a machine learning model to get more transactions. Thank you so much. I would like to thank again everyone for attending this session. Thank you, Mr. Al-Hasan. Sure, Rawan, it's always a pleasure. Yes, hello, can you hear me? Yeah, yeah, we do hear you and see you. I can hear you, yes. Yes, so I would like to thank everyone for attending this session. Thank you, Mr. Al-Hasan. From App Lab, thank you, Dr. Ayman from Ahmad Al-Khalid University. Excuse me. Thanks. Thanks a lot for the opportunity, Rawan. It was a pleasure. Thank you. And thank you again for Dr. Ayman, for Dr. Munir and for Mr. Yazan Asafi. I would like to take the opportunity to invite everyone to attend our second session tomorrow as well in the collaboration with Ahmad Al-Khalid University. We will be covering a very interesting subject under a blockade with two different sessions. I encourage everyone to attend this session as well. Thank you so much and we look forward to see you all tomorrow. Thank you. Thanks a lot. Wish you a nice day. See you. Thank you. Thank you. Have a good day. Thank you so much.