Lecture 01 - Overview of Cloud Computing

Summary

In this introductory lecture on cloud computing, Dr. Raji Mishra from the Indian Institute of Technology Patna provided a detailed overview of the evolution and current state of cloud computing. The lecture covered the foundational aspects of cloud technology, its history from distributed systems, and the economic projections that highlighted its rapid growth and adoption. Key features and benefits such as scalability, cost-efficiency, virtualization, and massive scale infrastructure were discussed. Dr. Mishra also addressed the architectures of private and public cloud services, the role of virtualization, data-intensive computing, and the significant shift towards a cloud-focused paradigm. The lecture concluded with an analysis of the cost-benefit between outsourcing to cloud services and owning infrastructure, providing a thorough understanding of cloud computing's implications on today’s technological landscape.

Highlights

• Dr. Raji Mishra provides a comprehensive overview of cloud computing's evolution from its distributed system roots. 🌏
• Predictions from 2009 and 2010 indicated exponential growth in cloud revenues and adoption. 📈
• Cloud computing supports scalability and cost-efficiency, enabling rapid deployment of applications. 🚀
• Virtualization plays a key role, allowing for multiple virtual machines to operate on single physical servers. 🕹️
• Public and private clouds offer different service models with unique advantages and accessibility options. 🔑
• Data-intensive computing has become critical, emphasizing the importance of cloud-based resources. 💾
• Cost analysis reveals when it's financially viable to use cloud services versus owning hardware infrastructure. 📊

Key Takeaways

• Cloud computing has evolved from distributed systems, offering scalable and cost-efficient solutions. 🌥️
• Virtualization is central to cloud technology, allowing for efficient resource allocation. 🖥️
• Massive scale infrastructure supports extensive data management and application deployment. 📊
• Cloud computing offers 'on demand' services that can significantly reduce operational costs. 💸
• There are different cloud models: public clouds for widespread access and private clouds for internal use. ☁️
• Choosing between cloud outsourcing and owning infrastructure depends on usage and cost analysis. 💡
• Cloud advancements have shifted the tech paradigm towards ubiquitous and flexible computing resources. 🔄

Overview

Cloud computing, as articulated by Dr. Raji Mishra, has grown dramatically from its initial roots in distributed systems. This transformative technology now underpins much of the global computing infrastructure, offering unparalleled scalability and flexibility. Dr. Mishra explains how cloud services have been developed to cater to dynamic workloads through virtualization and extensive data center networks, presenting a flexible solution for different business needs.

In highlighting the various components of cloud computing, Dr. Mishra emphasized the importance of virtualization technology, which allows multiple virtual machines to run on a single physical machine. This capability is crucial for the efficient management of computing resources and supports the scalable nature of cloud computing. He also described the cloud's capacity to support large-scale data management, meeting the demands of today’s data-centric applications.

Furthermore, the lecture provides insights into the decision-making process for companies considering cloud services versus traditional infrastructure investment. Dr. Mishra discusses the economic implications and efficiencies of adopting a cloud strategy—outlining how factors like cost, scale, and flexibility should inform whether a business should opt for public cloud services, establish a private cloud, or build its own infrastructure.

Chapters

• 00:00 - 10:00: Introduction and Overview of Cloud Computing This chapter introduces the concept of cloud computing. Dr. Raji Mishra from the Department of Computer Science and Engineering at the Indian Institute of Technology Patna presents the overview. The chapter aims to provide a brief introduction and set the stage for more detailed discussions on the subject in subsequent sections.
• 10:00 - 20:00: Evolution and Growth of Cloud Computing This chapter discusses the evolution and growth of cloud computing, starting with today's landscape and then looking into its development from earlier distributed systems. It covers the brief history and the current demand for cloud services, focusing on data center features and distinguishing between private and public cloud offerings. These topics are all part of the lecture's comprehensive exploration of cloud computing.
• 20:00 - 30:00: Virtualization and its Role in Cloud Computing This chapter discusses the evolution of cloud computing, emphasizing its significant growth forecast made in 2009. Predictions indicated that cloud computing revenue would exceed $150 billion by 2013 and comprise 19% of IT spending by 2015, as per IDC's forecast. The chapter likely examines these forecasts' accuracy and the impact of such growth on the technology landscape, potentially discussing the role of virtualization in enabling this expansion.
• 30:00 - 40:00: Cloud Infrastructure and Services The chapter discusses the growth of cloud infrastructure and services, highlighting predictions made by Forester regarding the increase in spending on IT cloud services. The forecast mentioned that spending would triple over five years, reaching $42 billion, and emphasized a trend in the industry growing from $4.7 billion in 2010 to $241 billion by 2020, indicating significant hype and movement towards cloud solutions.
• 40:00 - 51:00: Public vs Private Clouds and Conclusion Cloud computing has significantly evolved due to substantial investment from companies and governmental adoption, leading to the current expansive and scalable cloud landscape over the Internet.

Lecture 01 - Overview of Cloud Computing Transcription

• 00:00 - 00:30 [Music] overview of cloud computing I am Dr Raji Mishra as a professor Department of computer science and engineering Indian Institute of Technology Patna in this particular overview what we are going to cover is about about the brief introduction of a
• 00:30 - 01:00 cloud computing that becomes the background with the focus on aspects of today's Cloud then we will talk about the cloud computing from previous generations of distributed systems brief history current demand of cloud data center features of cloud computing private and public Cloud are some of the topics which we are going to discuss in this particular lecture as part of the or
• 01:00 - 01:30 view of cloud computing now let us trace the hype of cloud which has grown to this stage of today's Cloud now forecasting agencies gner in 2009 forecasted that cloud computing Revenue will soar faster than expected and exceed $150 billion by 2013 and this will represent 19% of it spending by 2015 IDC forecast in 2009
• 01:30 - 02:00 that spending on it cloud services will triple in next five years and reaching 42 billion uh dollars Forester in 2010 forecasted that cloud computing will go from $4.7 billion in 2010 to 241 billion in 2020 now based on these kind of Hypes there is a complete Trend that industry
• 02:00 - 02:30 is spending so much of money in it cloud services and which has resulted in the growth of cloud computing companies and even state governments are also started using the cloud which has resulted into this today's Cloud now let us see about the cloud that will give a scalable Computing over the Internet so evolutionary changes
• 02:30 - 03:00 have occurred in distributed and cloud computing over last 30 years which are now driven by the applications and variable workloads and the large data sets these are some of the challenges or the requirements of today's Cloud which has changes from distributed and in the cloud computing scenario now these evolutionary changes uh have been made in the machine
• 03:00 - 03:30 architecture operating system network connectivity application workloads so therefore distributed computing system uses now multiple computer to solve large scale problems and that to or the internet so distributed computing becomes data intensive and network Centric over this particular internet based scalable Computing now the emergence of computing Cloud thereafter increases in instead demands High
• 03:30 - 04:00 throughput Computing systems built with distributed computing principles so high throughput Computing appeared appearing as Computing clusters service oriented architectures computational grids peer topers networks internet clouds and future internet of things so these are some of the precursors from distributed computing to the current stage of
• 04:00 - 04:30 scalable Computing over the Internet which is also now has gone into grown into the cloud computing par L now with this little background how from distributed computing it has grown to the current cloud computing now this particular Computing cloud computing is being provided by different companies and they are called Cloud providers so the
• 04:30 - 05:00 prominent Cloud companies are Amazon web services Microsoft Azure Google compute engine or app engine right scale Salesforce EMC gig space Tangen Oracle VMware Cloud era and 100 more companies are having the presence as the cloud providers now with this cloud computing the customers often save time and money both for example with AWS a new server
• 05:00 - 05:30 can be up and running within 3 minutes compared to even 7 and a half weeks to to purchase and deploy the server internally in the form of a 64 node Linux cluster can be uh can be uh running in a online mode in a 5 minutes internally with the online services this will reduce the it operational cost by roughly 30% of his spending so the private cloud of virtual servers inside
• 05:30 - 06:00 data centers has saved nearly crores of rupees annually because the company can share the computing power and storage resources across the servers thereby uh there is a possibility of saving time and money both from the customers perspective hundreds of startups can harness this large Computing resources without buying their own machines in this way
• 06:00 - 06:30 customers save time and money both now this particular cloud computing which often allows sharing of computing resources across multiple customers inuring into resulting into the saving of time and money deals are is being provided with the help of a technology which is called the virtualization so virtualization has brought the benefits of the cloud
• 06:30 - 07:00 computing that is to the customers time and money and time both savings so let us understand about what you mean by the virtualization and and how is it uh giving up the cloud Technologies cloud computing Technologies so virtual work spaces that is an abstraction of an execution environment that can be made dynamically available
• 07:00 - 07:30 to authorized clients by using a well- defined protocols now it provides the resource quota that is in the terms of CPU memory share then software configuration now this particular virtual workpace is being also facilitated using virtual machines or VMS so VMS are nothing but an abstraction of a physical host machine and the hypervisor intercepts and emulates the instruction from VMS
• 07:30 - 08:00 and allows the management of virtual machines now important hypervisors are vmw Zen Etc and it provides an infrastructure which is supported by an apis and plugin to the hardware or the software and source so let us see here in this particular in this is stack here you can see this is the hypervisor
• 08:00 - 08:30 and this is the hardware and this entire thing is called virtualized stack up to this point now when you talk about the virtual machines so this part which contains the operating system and the applications running is called the virtual machines so virtual machines are this particular box so you can see here you
• 08:30 - 09:00 can form three different boxes so you can say that there are three different virtual machines are running on top of the hypervisors this is the hypervisor so if you see this particular virtual machine so it becomes an abstraction of the physical host machine so you can see that this particular application is running on its own
• 09:00 - 09:30 operating system the other application is running on another host operating system and all these three different virtual machines will become an abstraction of a physical host machines now here you can see that to support three different abstraction of physical machines it requires something which is called an hypervisor so hypervisor intercepts
• 09:30 - 10:00 and emulates the instructions which are coming from these virtual machines like here for example there are three different abstractions this is one this is second one and this is third one so they are being intercepted by the hard by the hypervisors and then allows the management of these virtual machines sharing this particular Hardware uh across these virtual machines so this
• 10:00 - 10:30 particular prominent hypervisors are VM Zen Etc now this particular uh hypervisor they provide the infrastructure apis and that can be plug into the hardware and support system now let us go in more details over here is the that virtual machine
• 10:30 - 11:00 technology allows multiple virtual machines to run on single physical machine here that example is shown work in this particular slide that you can see that that this is the hardware this is one piece of Hardware there is one physical machine or you can also say this is a single physical machine and this particular single physical machine is being shared by by three virtual machines that you can see this is is the virtual machine
• 11:00 - 11:30 number one this is virtual machine number two and this is another virtual machine 3 so there are three different virtual machines they are sharing the same hardware and that is being facilitated with the virtual machine technology which is called virtual machine monitor or it is also called the hypervisor now let us see what are the advantages uh here of virtual machines so virtual machines if you see in the previous
• 11:30 - 12:00 figure virtual machines they have their own guest operating system and this guest operating system runs multiple applications so advantage of virtual machine is that it will run operating system where physical Hardware is unavailable so only the virtual form of uh Hardware is being uh presented or let us say that many virtual machines which
• 12:00 - 12:30 runs their own operating system without any physical Hardware is allowed by the virtual machines now advant another advantage of virtual machine is easier to create new machines and backup machine Etc another Advantage forms about software testing using clean installed operating system and software emulate more machines Than Physical physically available time share lightly loaded systems on one host debug
• 12:30 - 13:00 problems suspend resume problems easy migration and run Legacy systems now with this little background of a cloud and the use of hypervisor now let us understand what is the cloud so the advances in virtualization makes it possible to see the growth of cloud as a new Computing Paradigm so dramatic difference between develop in the software for billions to use as a
• 13:00 - 13:30 service versus Distributing software to run on their PCS was being realized with the help of a cloud so let us Trace back the history to understand what is a cloud 1984 John goge from Sun microsystem said for the cloud that network is the computer in 2008 David Patterson of us Barkley said the data center is the computer recently Rajkumar baa of melbour said the cloud is the
• 13:30 - 14:00 computer now some people view clouds as the gits or clusters with changes through virtualization since the clouds are anticipated to process huge data sets generated by the traditional internet social networks and future iot so the cloud can be a single side Cloud which is also known as the data center often comprises of a Computing noes often grouped into the racks that is
• 14:00 - 14:30 shown by this red arrows then in a single side Cloud not only compute nodes which are grouped into the racks but it also has the switches which can connect across different racks that you can see the switch which is connecting three different racks and which is also called top of the rack switch then it has the topology that is Network topology of all that networking devices that is top the rack switch is all connecting the
• 14:30 - 15:00 servers or the racks are often connected together with the help of a code switch then it has the storage of that is the back end nodes connected to the network then it has a front end for submitting the jobs and receiving the client requests and this entire thing is often called as three tier architecture and it runs the software services that was a single site cloud in in the
• 15:00 - 15:30 cloud and it is also called the data center now another form another thing is there inside cloud or to understand the cloud is that it's a geographically distributed Cloud which consist of multiple such site or a multiple single s side cloud and each site perhaps with a different structure and the services together it's called geographically distributed Cloud now cloud computing overlaps with
• 15:30 - 16:00 distributed computing as you know that distributed systems are a precursor to the cloud so they often uh overlaps with the cloud computing overlaps with the distributed computing in that sense so distributed computing in distributed system consist of multiple autonomous computers having their own memory communicating through messages that is the simple form of distributed system which gives it into the distributed
• 16:00 - 16:30 computing Paradigm whereas the cloud computing the clouds are built with the physical or the virtualized resources or the large data centers that are nothing but a distributed system so cloud computing is also considered to be a form of utility computing or the service Computing so if you can see the difference that distributed computing is nothing but the distributed systems where can see that the aut autonomous computers which are connected
• 16:30 - 17:00 with the network having their own memory they often communicate with each other using the messages or a message passing so message passing system uh um which is being used to communicate across the multiple autonomous computer forms a distributed computing similarly as far as the cloud computing is concerned so the clouds can be built with the physical or the virtualized resources or the large data centers that
• 17:00 - 17:30 are distributed uh systems so this particular entire distributed system when it is virtualized will give a virtualized resources and that are nothing but uh they are built with a large data centers and often virtualized so cloud computing is also considered to be in that way the form of utility computing or the service Computing so a cloudy
• 17:30 - 18:00 history for a cloud computing let us Trace back the history of the time it started with the First Data Center and then you know that it's a large system uh in 1950s and they were also called initially it's a data center then then comes the time sharing systems in 1970s 1980s it it it it forms into the GR and the cluster Computing and then in 201 uh2 around
• 18:00 - 18:30 that time it became the cloud data center uh before that it was also called as a peer-to-peer system they're all precursors to the cloud data center or peer-to-peer system grid and clusters time share systems and the First Data Centers and again the wheel has come to the same point where it became the cloud data
• 18:30 - 19:00 centers now if you see the scalable Computing trends that is the technology so there you can see that the storage is doubling every 12 months on the same cost bandwidth every 9 months and the CPU compute capacity every 18 months and this is called the moods law indicates that the processor speed doubles every 18 months gilder's law indicates that the network bandwidth has doubled each year in the past then and now the
• 19:00 - 19:30 bandwidth earlier in 1985 was mostly 56 kilobytes Nationwide and now in 2015 you can see that this particular backbone has become terabyte terabits per second from kbps similarly it has the disc capacity that is today species have uh terabytes for more than 1990s super computer so the trend towards utility
• 19:30 - 20:00 computing is that aiming towards autonomic operations can be self-organized to support Dynamic Discovery major Computing paradigms are composable with quality of service and service level agreements in 1965 mit's Fernando forbat of multics operating system and envisioned a Computing facility operating like a power comp or a water company that is the utility
• 20:00 - 20:30 computing plug your Thin Client into the Computing utility and play intensive compute and communication application utility computing focuses on the business model in which the the customers receive the Computing resources from a paid service provider so the characteristics of cloud computing uh some common characteristics is that it is having a massive scale it is having the homogeneity and then virtualization low cost software
• 20:30 - 21:00 resilient Computing geographic distribution service orientation and advanced security NOW essential characteristics is that on demand self-service broad network access rapid elasticity resource pooling and measured services are some of them so now let us go inside of all that they are called the features of today's cloud and they are all divided into four different
• 21:00 - 21:30 categories to understand all the features of today's Cloud the first important feature is called the massive scale in today's Cloud so what do you mean by the massive scale it is nothing but a very large data centers contains tens and thousands and sometimes hundreds of thousands of servers and you can run your computation across as many as server as you want as as many as server your application will scale and that that huge infrastructure which is
• 21:30 - 22:00 providing the application which requires the huge infrastrcture together is called the massive scale in today's cloud is important feature another important feature is the second important feature is called On Demand access pay as you go no upfront commitment that is anyone can access it third important feature of cloud is data intensive nature what megabytes has now become terabytes and petabytes and
• 22:00 - 22:30 databytes for example daily locks forensics web data are a very large or a they can also be rated as the big data so this kind of Big Data are huge amount of large scale data and requires the Computing called Data intensive nature of computation is one of the important feature of today's Cloud fourth important feature is that new Cloud programming Paradigm that is
• 22:30 - 23:00 map reduce Hardo no SQL Cassandra mongod and many others so combination of one or more of these gives rise to novel and unsolved distributed computing problems now let us understand about the massive scale from the cloud providers perspective so Facebook you can understand the Facebook in 2012 has
• 23:00 - 23:30 180k number of servers similarly Microsoft uses 110,000 machines at four different sites to run the application that is called Microsoft Cosmos similarly Microsoft Bing requires 80,000 running systems Yahoo in 2009 uses 4,000 different clusters split into
• 23:30 - 24:00 4,000 different clusters of 10,000 100,000 machines AWS ec2 that is in 2009 uses to run on 40,000 Machines of eight course similarly eBay in 2012 uses to run 50,000 machines HP 2012
• 24:00 - 24:30 uses 380,000 180 different data centers Google has a lot so this is the size this is called massive scale if you look inside the data center from that perspective of scale or the size or the massive scale that you can see that these are the figure you shown they are nothing but they are called the racks the room which is filled with the
• 24:30 - 25:00 racks and rack is filled with such so many number of servers and all are connected with a three tier hierarchy Network topology and they're all connected together can see over here now this entire massive data center having hundred thousands of computer requires huge power and energy so they are power usage efficiency is a matric uh
• 25:00 - 25:30 which will tell pu e pu is called power usage efficiency so power usage efficiency is total facility power divided by an IT
• 25:30 - 26:00 equipment power so low is good so Google says that it is achieving 1.11 uh that is power usage efficiency similarly water usage efficiency water is used to cooling uh for cooling purposes so so the efficiency here is water usage efficiency is nothing but the annual water uses divided by it equipment energy so a lot of new lot of cooling Technologies are involved in the data center now coming up the second
• 26:00 - 26:30 important feature of a cloud today's cloud is called On Demand access so on demand access uh is also called as a service classification so as a service classification here it is shown as the cloud application that is uh starting with the cloud infrastructure as a service if you see Cloud infrastructure as a service called IAS so Cloud infrastructure as a service that is being provided in the form of
• 26:30 - 27:00 virtual machines and this is be this is being achieved with the help of having lot of servers with the cloud providers storage load balancer and network together it it provides virtual machine and this is called Cloud infrastructure the second sophisticated service then infrastructure service called platform as a service that is
• 27:00 - 27:30 p so there it provides the development tools or the environments instead of uh dealing with the the virtual machines now Beyond this there is another type of as a service called software as a service so there the application services are being provided and that is not even dependent upon
• 27:30 - 28:00 virtual machines or a platforms but directly the services for example Gmail virtual desktop CRM and so on so you can see that when you say an on demand access so you have to now compare it in terms of buying or renting as a service like cloud so AWS is one
• 28:00 - 28:30 such service which is called elastic compute Cloud a few cents to a few dollar per hour CPU usage is being measured similarly AWS simple storage service S3 is a few cents per GB month Hardware as a service infrastructure as a service that is get access to the flexible computer storage infrastructure virtualization is
• 28:30 - 29:00 is the technology which gives which provides infrastructure as a service examples are Imagine web services AWS EC provides two different type of infrastructure service ec2 is a compute plastic compute cloud is a compute service and S3 is a Simply Storage service that is a storage as a service platform as a service get Flex get access to the flexible Computing and storage infrastructure coupled with the software platform example is Google's
• 29:00 - 29:30 app engine which provides the platform as a service for python Java and go platforms software as a service SAS get access to the software Services when you need them and subsume the software as a service architecture examples are Google doc or an example of a software service now uh if you see the uh descript service versus description what you will find is that infrastructure focus is their hosting
• 29:30 - 30:00 storage and platform that is hosting means that physical data centers which are run by the presence of different companies like IBM HP um Google Microsoft and so on storage means data storage or a cloud-based Nas platform is the platform such as cloud-based platforms provided by virtualizations as we have already told application focused Services
• 30:00 - 30:30 applications and development are uh we have already seen that in the previous now the next feature of a cloud is data intensive Computing so computation intensive Computing we have already seen earlier days like called MPI based high performance and grid computing and super computer was called uh often the computation
• 30:30 - 31:00 intensive Computing but now we have moved to another era of computing which is called Data intensive nature of computation typically it stores the data at the data center and use the Computing notes nearby so why the data is too huge it cannot move where the compute notes are but rather the compute notes if it is being available near to the story stage places then it is a possibility
• 31:00 - 31:30 and it is called Data intensive nature of computation so the comput not runs the computation Services now in a data intensive Computing the Focus Shift from computation to the data now CPU utilization no longer the most important resource matx instead of here the input output is also very important such as disk and the network finally the next Paradigm next feature of a cloud computing today's cloud is called New Cloud programming Paradigm
• 31:30 - 32:00 easy to write and run highly parallel programs in a new Cloud programming paradigms such as Google provides map ruce and szal and Amazon provides elastic map ruce a pay as you go by model now Google Map ruce provides indexing as a chain of 24 map rued jobs around 200,000 jobs jobs they are being processed with 50 paby per month in 20
• 32:00 - 32:30 2006 Yahoo has Hadoop and pig together Facebook has Hadoop and ha and no SQL is my SQL is an industry stand cendra can see uh provides 2,400 times faster using this new Cloud programming paradigm now there are two categories of clouds
• 32:30 - 33:00 first is called public Cloud the other is called private Cloud private clouds are accessible only to the employees for example a popular vendors for creating the private clouds or VMware Microsoft Azure eclips public Cloud are being provide the service to anyone uh to who can be a paying customer examples of a public cloud services include uh Amazon AWS ec2 Google app
• 33:00 - 33:30 engine Gmail Office 365 and so on so if you are starting a new company or a startup you should use a public cloud or purchase your own cloud so this particular decision has to be made Single side Cloud to Outsource or to own let us understand this particular decision or a question consider a mediumsized
• 33:30 - 34:00 organization that wishes to run a service for a months now that service which mediumsized organization wishes to run requires 128 servers for that application and it requires 1024 course and 524 terabytes so let us see that if you if that organization decides to Outsource
• 34:00 - 34:30 via AWS service let us consider the monthly cost now for the storage AWS charges S3 cost as .12 per gigabyte month and ec2 cost is10 per CPU hour so let us see about the storage how much costs for requirement of the
• 34:30 - 35:00 storage that is 524 tabes multiplied by 1,000 because it is12 rate is .12 per gigabyte month it comes out to be $62,000 now uh calculating the CPU hour now the CPU time
• 35:00 - 35:30 is given as .10 and it requires 124 different cores and it requires 24 multiplied by 30 24 hours and 30 days it comes out to be do 136,000 that is if it Outsource on a monthly basis similarly to own that that means if the organization is going to purchase it
• 35:30 - 36:00 what is the monthly cost works out to be for the storage is $ 34900 $1,000 divided by m months that comes out to be total of $555,000 divided by m mons plus 7.5 key that includes the system Administration cost considering uh that the split for
• 36:00 - 36:30 Hardware power and network uh and threee lifetime of the hardware now let us see calculate the bra even analysis more preferably to own if 349 K dollar divided by month is less than 62,000 for storage and
• 36:30 - 37:00 for uh and similarly for uh this one um CPU infrastructure that is 1,555 K divided by m + 7.5k if it is less than $136,000 overall so Brack even comes out to be if m is that is m is greater than 5.55 months for storage and M is greater than 12 months overall so that means if medium siiz organization wishes to use
• 37:00 - 37:30 the storage for more than 5.55 months then it is better to own that it will be uh Break Even cost analysis says it similarly if M that is the number of months is more than 12 then uh it says that overall uh system has to be now better uh overall system has to be now procured uh and use it so most of the
• 37:30 - 38:00 startup companies uses the cloud a lot because it is uh bra one point comes out to be less than a year and the cloud providers then benefit monetarily moves from the St so conclusion we have covered the following Topic in this lecture we have given the overview of cloud computing then cloud computing versus the previous generation of Cloud that is nothing but a distributed computing we have given brief history
• 38:00 - 38:30 and the current demand of cloud data centers various features and characteristics of cloud computing on demand cloud computing with private and public clouds thank [Music] you
• 38:30 - 39:00 [Music] [Music]