Cloud Economics

Summary

The talk explores the economic considerations of cloud computing and why organizations might switch to cloud infrastructure. Economic factors such as cost benefits, utility pricing, scalability, and statistical multiplexing make cloud adoption appealing. Emphasis is placed on the balance between in-house infrastructure versus cloud, and factors like peak demands, variability, and associated costs when transitioning to or using cloud services. Additionally, the discussion covers the risk-return trade-off, utility premium, and the economic scale in providing resources efficiently.

Highlights

• The cloud provides an economic advantage through pooled and standardized resources. 📊
• Statistical multiplexing benefits are crucial for understanding cloud economics. 📈
• Cloud adoption requires consideration of peak demand and infrastructure balance. 🏗️
• The risk-return trade-off is significant when transitioning to or utilizing cloud services. 📊
• Utility pricing models such as pay-as-you-go are essential in cloud economics. 💸

Key Takeaways

• Cloud adoption is largely driven by economic considerations and can offer significant cost benefits. 🤓
• Using the cloud means leveraging pooled, standardized resources, creating efficiencies through statistical multiplexing. 🌐
• Organizations need to balance between in-house infrastructure and cloud platforms based on peak demands and costs. ⚖️
• Understanding utility pricing and on-demand services is crucial for deciding whether to move to the cloud. 💡
• The coefficient of variation helps assess the risk-return trade-off in cloud investment. 📉

Overview

In the world of cloud computing, economics plays a pivotal role in decision-making. The transition to cloud services allows organizations to access pooled, standardized resources, which offer benefits of statistical multiplexing. This is particularly significant when balancing between keeping in-house infrastructure and opting for cloud solutions. Merging in-house and cloud capabilities helps meet peak demand while controlling operational costs.

At the heart of cloud economics is utility pricing, which refers to the 'pay-as-you-go' model. This model ensures that organizations only pay for the services they use, making it potentially cheaper than maintaining in-house systems. The utility premium—how much more it costs to use cloud services compared to standard methods—plays a crucial role in these economic considerations. Understanding these pricing strategies can guide organizations toward more efficient resource use and cost management.

Additionally, the coefficient of variation becomes an important metric when assessing the risk associated with cloud investments. It’s a measure of smoothness in demands, and lower CV indicates lesser risk. The talk explored how organizations can use these economic principles to optimize resource allocation, enhance performance, and achieve a better return on investment in cloud services.

Chapters

• 00:00 - 00:30: Introduction to Cloud Economics This chapter serves as an introduction to the economics of cloud computing. It begins with a continuation of previous discussions on cloud computing and transitions into the exploration of the economic factors and considerations pertinent to cloud technologies.
• 00:30 - 01:00: Cloud Viability and Business Considerations The chapter explores the concept of cloud computing and why businesses might choose to adopt it. It explains that while cloud computing isn't entirely new, it offers a different model by providing applications as services. The discussion revolves around the viability of moving to the cloud and the factors that make it beneficial for businesses. The narrative seeks to answer whether it is always advantageous to migrate to the cloud or if there are specific conditions under which cloud adoption becomes more favorable.
• 02:00 - 03:00: Common Infrastructure and Statistical Multiplexing The chapter titled 'Common Infrastructure and Statistical Multiplexing' focuses on decision-making processes regarding whether to adopt cloud computing or maintain in-house infrastructure. It explores how to balance between these options and examines the economic factors that make cloud computing a viable choice for certain operations and situations.
• 04:00 - 05:00: Peak Load Considerations The chapter 'Peak Load Considerations' addresses the various factors organizations or individuals should keep in mind when transitioning to cloud services, whether partially or fully. It touches upon Service Level Agreements (SLAs) and other cloud-related issues discussed in previous lectures, highlighting that even if these factors are managed well, organizations need to be aware of potential limitations and challenges associated with cloud adoption.
• 06:00 - 07:00: Cost of Over-provisioning and Location Independence This chapter discusses the economic aspects of cloud computing, particularly focusing on the cost implications of over-provisioning and the benefits of location independence. It highlights considerations businesses should take into account when determining the optimal use of cloud resources, including whether to always remain on the cloud or to do so at specific times. The chapter aims to assist those interested in leveraging cloud technologies by re-examining various cloud properties, such as the concept of a common infrastructure.
• 08:00 - 10:00: Utility Pricing and On-Demand Resources The chapter discusses the concept of pooled standardized resources and their benefits, particularly through statistical multiplexing. It explains how service providers manage a pool of standardized resources and highlights the advantages of multiplexing tasks statistically. This method allows for efficient integration and optimization of resources, bringing about benefits for both service providers and users.
• 12:00 - 15:00: Economies of Scale and Overhead Cost Reduction In the chapter titled 'Economies of Scale and Overhead Cost Reduction', the discussion centers around the inefficiencies of using systems in terms of workload management. It highlights how a significant portion of resources, sometimes between 15-20%, remain underutilized. This underutilization prompts a reconsideration of the necessity to invest in a large number of systems, such as buying ten or even a hundred systems, as most of the time, the workload requirements do not justify such extensive resources. The argument suggests that economies of scale can be achieved by optimizing the use of existing systems rather than unnecessarily expanding the number of systems based on peak workload scenarios.
• 15:30 - 18:00: Coefficient of Variance (CV) This chapter discusses the concept of Coefficient of Variance (CV) in the context of managing resources for peak workloads in an educational institution, like IIT Kharagpur. The example given describes the setup of a lab for a specific department and year of the M Tech program, where there are 50 seats. The aim is to manage these resources effectively, particularly during peak times, by ensuring there is adequate equipment and power to support individual systems in the lab.
• 20:30 - 23:00: Latency and Location Independence The chapter discusses the concepts of latency and location independence in a technological and educational context. It explores the idea of having systems (such as computers) on standby to handle unexpected increases in demand, suggesting that additional systems (around 5-10%) should be available as a backup. However, in reality, the number of users or students enrolling may be less than expected, leading to underutilization of these standby systems. This underutilization highlights the challenge of efficiently managing resources to meet fluctuating demands.
• 27:00 - 29:00: Utility Pricing and Real-World Scenarios The chapter titled 'Utility Pricing and Real-World Scenarios' discusses the usage of resources during lab hours, specifically focusing on a daily eight-hour utilization period. It highlights how systems are evaluated individually with considerations such as processor and memory to handle peak loads. These considerations often lead to oversizing to accommodate high demand periods, emphasizing the criticality of having adequate resources.
• 33:00 - 35:00: On-Demand Services and Penalties The chapter discusses the management of a server that caters to multiple users, particularly focusing on handling peak loads. It describes a scenario where numerous users may attempt to access the server simultaneously, highlighting the challenges associated with network capacity. The example given entails a typical 24-port network, which can support up to 24 connections even at high bandwidths like 100 Mbps. The explanation avoids delving into the complexities of binary calculations, focusing instead on the practical impact of such a network setup and the potential penalties of exceeding its capacity.
• 35:00 - 37:00: Assignment and Conclusion The chapter titled 'Assignment and Conclusion' discusses data provisioning and network architecture. It addresses the concept of over-provisioning versus statistical demand, illustrating the challenges of providing sufficient uplink bandwidth without unnecessarily exceeding practical requirements. The discussion centers around the problem of network uplinks handling traffic, where a two-point-four-gigabyte requirement may be mismatched with a one-gigabyte uplink, leading to a consideration of whether or not to provision a higher capacity, factoring in the statistical peak times and usage patterns.

Cloud Economics Transcription

• 00:00 - 00:30 hello so we will continue our discussion or lectures on cloud computing today we will take up a topic which is need to we need to look at that what is this economy behind cloud
• 00:30 - 01:00 computing why people will go for ah this cloud computing type of things right what it is it is not like that we are getting a something totally new so it is it is only that the different type of application etcetera now we are getting as a services so it is what makes this or what will make this viable is it always going to cloud is beneficial or where
• 01:00 - 01:30 how to decide that whether i need to go to cloud or whether i need to buy something in house and how much how to balance between in house infrastructure and provisioning of the cloud so in order to do that we will try to see some basic phenomena or basic economic point which is makes cloud viable at which type of operations or which type of situation
• 01:30 - 02:00 type of things so we try to more try to look out that when we when a organization or individual especially organizations one switch over cloud partially or fully what are the consideration it should keep in mind we have seen slas we have seen they are there are other issues that even during cloud our initial lectures that there are some limitation there are some issues with the cloud but keeping all that even all those things are running fine whether it is
• 02:00 - 02:30 economically to be on cloud always or at times or at what times and type of things and what should be the business consideration so we will have a brief discussion on that so those of you are interested in working on this line can basically now leverage on this type of work so ah from the economic point of view if we if we see that what are the what are the different cloud properties relook at the properties ah one is the common infrastructure
• 02:30 - 03:00 so pooled standardized resources with benefits generated by statistical multiplexing right so that is important first of all it is a pooled standardized resources so at the service provider end it is a pool of resources which are standardized how they can be integrated etcetera right what is the benefit from from the thing because of multiplexing tasks right so statistically multiplexing benefits statistical multiplexing that means i have some hundred
• 03:00 - 03:30 systems right so it is highly underutilized even fifteen twenty percents are not utilized of my own own workload so or from some workload so if i have multiple workload i may with the same type of system i can go up into the things what i am thinking that why i have purchased this some ten or hundred systems because considering most of the cases we have
• 03:30 - 04:00 considering a peak workload right at peak time i will be require all those things right say for example iit kharagpur in a particular department a particular lab for a particular year say ah m tech first year the number of seats are say fifty for a particular department and what we expect that fifty is fifty percent we will do work on fifty individual system we purchase a lab of lab we basically set up a lab of fifty systems adjoining we same time we have to go for power at the same time
• 04:00 - 04:30 we have to go for ac and we feel that if there is a system goes out and type of things keep another five ten percent another five system on a standby so having fifty five systems but it may so happen that that dr the recruitment ah the number of students joined the course may be less than fifty so i have any surplus power even they joined the courses right this fifty system are again underutilized not may not be utilized all
• 04:30 - 05:00 the things during the lab hours may be ah daily eight hours it is utilized otherwise things are not there right one is looking at the peak load so but whenever i consider the system individually i have taken this processor memory etcetera thinking this lab assignment will be leading up to that level so that means at the peak load right so so this consideration may be or at many times i do lot of over sizing of the things and if i have is it becomes more critical when
• 05:00 - 05:30 i have a server a single server a server which is catering to number of users and then i think that all the user will jump into the thing right and then the peak load will be there right if those who are working in the networking you might have seen that typically ten twenty four port networks which right how many people can connect twenty four even hundred mbps line so it is twenty four roughly twenty four into hundred if we not go to the netegrity of this ah binary thing
• 05:30 - 06:00 so it is approximately two point four gigabytes but the switch uplink will be maybe one gigabyte so it is some sort of a blocking architecture but if i give a uplink of two point four or three gigabyte then it is over provisioning right i it is all the people are coming statistically at the same peak time may not be either there so it statistically we need to look at that whether it is viable to provision such a higher thing when you provision higher thing it involves
• 06:00 - 06:30 lot of costing and other things into the things maintenance of the thing there are things of adjoining other accessories etcetera even the cost of ah the equipment goes up and so and so forth so that is one thing another is the location independence that is another property of cloud like ubiquitously available ah meeting performance requirement with benefits deriving from latency reduction and user experience in enhancement so this is a location independence
• 06:30 - 07:00 property so whether it is economically how to make use of that online connectivity as an enabler of other attributes ensuring a service cost cost and performance impact on the network architecture etcetera so i should have a online always connectivity so there is another factor so these are the different factors which may ah force us to look at the different things there are other two direct economic factor one is the utility pricing right i pay as
• 07:00 - 07:30 you go model i you i pay per unit things like electricity etcetera there is another thing called on demand resources so when i when i demand for the resources resources are provisioned right so it is on demand i demand for the resources are provided so scalable elastics resources provision and de provision without delay and cost associated with the change right so there cannot be delay or cost associated anything there may be cost factor rather there should not there should be minimal human on managing many managerial involvement so that
• 07:30 - 08:00 as as i go and go forth and things i may resources provisioning and de-provisioning so with this we want to ah try to find out this valuation of this or whether there is a economic point of view we try to look at little bit statistically very very little portion of they need to understand this issue one is economies of scale reduced overhead cost by our power through volume
• 08:00 - 08:30 purchase right one is that i want to one is the economy scale means i want to reduce this overhead cost right so what are the different overhead cost like if i buy a system it has overhead cost of ac it has overhead cost of power maintenance like ups or it has overhead cost of amc of the system right and there are several other of like human resources to maintain the thing etcetera so i want to reduce that overhead cost right
• 08:30 - 09:00 so buying the thing is still ok but maintaining the thing at times becomes over the years much costlier than the equipment itself right and there is a major problem of especially for in the computing world that after typically two to three years not more than definitely within five years that whole thing becomes obsolete the technology becomes obsolete the whole system power etcetera is no longer valid for no longer becomes viable for installing
• 09:00 - 09:30 new set of tools and software etcetera so that is a big problem there is a statistics of scale on the other end for infrastructure build on peak requirement like infrastructure build up peak requirement i think that i build an infrastructure that always all my students will be there in the class or all will register for the course and etcetera so multiplexing demand may help me higher utilization right so when i use things on multiplexing different demand may help me higher iteration lower cost for deliver resources then unconsolidated
• 09:30 - 10:00 work so it is a low lower cost per delivered resources then if there is a un consolidated workload so if it is a consolidated and lower cost will be there so for infrastructure build to less than peak so it is not peak less than peak here also multiplexing demand reduce the unserved demands right so multiplexing so there may be if it is a blocking architecture there can be there can be some of the things which are not served
• 10:00 - 10:30 so multiplexing may reduce this unserved this it is not like that it is good that one it is always biased to the one i go on some sort of a scheduling algorithm and go on serving low or loss of revenue or a service level agreement violation because sla violation means you need to pay out something ah sla violation payout so it may reduce lower loss of ah revenue and etcetera so both for peak peak infrastructure build on peak and non peak we have this sort of things another another term which comes not only for cloud for any
• 10:30 - 11:00 type of this type of things where this key is involved and you have to give services what you say coefficient of variance or commonly cv ah so it is not exactly covariance we are talking about coefficient of variance so a statistical measure of the dispersion of data in a data series around the mean right like a coefficient of variance is represented by the ratio of the standard deviation of the mean to the
• 11:00 - 11:30 mean so it is standard deviation or those who under sigma to me sigma by mu right here standard deviation by mean and it is useful statistics for comparing the degree of variation from one data series to another right so ah i can say that this whether the coefficient of variation of this data series is more than less than equal to other data series is this is a good measure to make look at it right rather even if the means are drastically different
• 11:30 - 12:00 then also we can compare two data series so how these two data series behavioral things are there the cv gives me idea to do that right so it is widely used or widely looked into in the investing world so in a investing world coefficient of variation allows you to determine how much volatile or risk you are assuming in comparison to the amount of return you can expect from your investment so this cv gives you a idea that how much
• 12:00 - 12:30 risk you are assuming in comparison to the amount of return you can expect from the investment so our y how much risk you are involving so this is important if you see it is also important in our this sort of scenarios also like we are we are basically involving some risk by leverage by putting my organizational from means infrastructure or on from the on premise infrastructure so cloud infrastructure so there is a risk of that if the infrastructure
• 12:30 - 13:00 is not available and so and so so it is not only infrastructure i want to mean to say that all type of services on the cloud thing so if the service is not available at it so how much risk i need to take on those things so in simple language lower the ratio of standard deviation to mean ah the better is your risk return trade off right so lower the ratio to standard deviation to mean return so the better is the risk so i
• 13:00 - 13:30 can have more smoothness into the curve that is important so it is a some sort of a measure of smoothness so if it is a very variable load or layer a lot of peak and non peak things then i am in much much ah bigger trouble in measuring that how things will be there if it is a smoothing out load then i am i am much in a better position to do that so coefficient of variation cv as you say that it is not similar as variance nor correlation coefficient
• 13:30 - 14:00 as we are mentioning ratio of standard deviation to sigma to absolute value of the mean mu or mu mod so smoother curves large mean for a given standard deviation so as we are telling the sigma by mu if it is a large mean then given standard deviation things will be not varying that much or a smaller standard deviation or a of a given mean that also things will be much under control so importance of smoothness a facility which fixed asset servicing highly
• 14:00 - 14:30 variable demand will achieve lower utilization then a similar one servicing relatively smooth demand right i have i am a service provider i have some hundred systems are my backbone and i I i know that the demand will be something smooth right then i can basically have a better management of the things or i have n number of systems so what will be the value of n is much easier but if it is a very much varying suddenly demand goes up hundred then ten etcetera
• 14:30 - 15:00 then i have a problem right similarly for any if you look if we look at our day to day life any shop shopkeeper the amount of provisioning you will do that you will keep this in its store this depends on the demand thing if the demand is something smooth it may vary that monday demand is different from tuesday demand different from week weekdays to weekends demand etcetera fine but he has a idea but it is totally random or ah you don cant do
• 15:00 - 15:30 anything then it is a very difficult to provisional things so thats the same thing holds there so multiplexing demand from multiple sources may reduce the coefficient of variation right that is the thing so now i cannot predict that who what are the different sources how things will be there but if i could have multiplexed the different demand from multiple sources so it may so happen that may may happen that this multiplex thing may have a give me a better cv right which may have a little smooth cv where the overall demand things are there
• 15:30 - 16:00 it is it is not like that all are going peak at the time all are coming down at the lower things but i have a multiplex type of things like like if we look at say x one x two x n are independent random variables of demand identical standard say for our argument sake they are having though they are independent and then but they have something identical sigma and mu so aggregate demand in case of mean some of the means in mu aggregate variance is n sigma
• 16:00 - 16:30 square so if we calculate that coefficient of variance it is one by root n cv right so if n increases i multiplex more than the one by root n decreases so i have more smoothing out this this coefficient of variation or moves smooth type of curve so adding n independent
• 16:30 - 17:00 demand reduces cv by one by root n so it is it becomes more smoothing out so penalty of insufficient excess resources go smaller right so what is happening if it is smoothing out then i dont have to plenty i have to keep more resources at my back end right it it it reduces right otherwise i dont know whether it is a ten ah demand of ten units or hundred units and etcetera and then i have to keep a track of hundred units at the back end like
• 17:00 - 17:30 aggregating hundred workload bring the penalty down by ten percent one by root over root over hundred is ten so it may bring down the whole thing by ten percent so aggregating multiple resources may allow me to have ah a reduced loading but what about the different workloads non negative correlation demands like if x and one by x sum of a random variable is one appropriate selection of the customer segments right is
• 17:30 - 18:00 another important thing right i say that i I have a computing infrastructure i select those customers some of them are active on day time some of them processings are active on the night time right so it is it is compensating right i if the both are working at the same time then the peak will go much but selecting that negative demands or the time in the timescale
• 18:00 - 18:30 i could have managed those thing with the same infrastructure so that is that selection of the customer base is the important things what sort of things will be there when i go on selecting as customers perfectly correlated demand if some of the things are perfectly coordinated the aggregate demand will be nx variations will be n square sigma square x and mean will be n sigma and standard deviation so and so forth coefficient of variation remains constant so it is perfectly correlated demand that this thing will be as a constant
• 18:30 - 19:00 there are third issue if all demands are coming at the peak at the same time all at the peak at the same time then i have a serious problem right then i have a congestion at the things all are all are demanding at the same time like if i say all classes are breaking at the hourly basis so at the hourly basis we have a huge ah demand for these rotor network right lot of vehicles like said studying for cycles to it said they are on the demand because
• 19:00 - 19:30 all classes are things so it is it is a peak coming to the thing at the same time like this then we have a problem so ah common infrastructure in our real world so correlated demands private means sides large sized providers can experience similar statistic scales so it is a more or less correlated demand independent demand midsize provider can achieve similar statistical economy to an infinitely large provider right available data on economy of scale for large
• 19:30 - 20:00 provider is mixed right because use of same quartz type of things that is commercial of the self systems and components locating near cheap power supplies that is one thing like if as the power supply is a major thing for the data centers they want to build a data center where the power supply will be much cheaper and nearby early entrain automation
• 20:00 - 20:30 tool third party parties takes care of it so there can there can be early entrant automation tools which the third party can be taking say value of there is a value of value corresponding to the location or value of location independence we use to go to the computers but the application services contains now come to us right so ah the pariah there is a paradigm shift like say yesterdays or we used to go to the system to work on it now the system power etcetera coming to my own next stop like so i have
• 20:30 - 21:00 a large pool of huge resources on my very thin ah system it can be a simple desktop laptop and type of things all are provisions here so through networks wired wireless satellite etcetera but what about latency so latency also a big thing right so human response is
• 21:00 - 21:30 ten to milliseconds ten to hundred milliseconds so latency is correlated strongly with the distance right more that distance more the network latency another type of latency more on the hops more on the failure rates and etcetera so though it also depends on the what sort of routing algorithms etcetera also coming into play so speed of ah anyway we know that speed of light in fiber so some particular one twenty four miles per milliseconds so if suppose i am searching something and it takes more than couple of seconds then we are not happy with that even a v o i p thing
• 21:30 - 22:00 if it is something delayed more than something two hundred nanosecond a two hundred millisecond second then it is very difficult to communicate over this voice over ip so there is a supporting of a global user base requires a dispersed service architecture so the architecture if i want to support if my provider is to support all the global user base then i have to have a appropriate distributed and dispersed architecture to do that and similarly the protocol so coordinate
• 22:00 - 22:30 and coordination consistency availability partition tolerance these are issues so and it has a direct implication on investment that what sort of investment we want to do we need to we like to look at another quickly another aspects of the thing what we say value of utility pricing right that how things will be priced how things will be there how i can
• 22:30 - 23:00 provision system and it may also help me to look at that whether it is useful to be go to cloud or having something at your own premises so economy of scale might not be very effective always means all all taking consideration right but cloud service dont need to be cheaper to be economical right so it is economic cop
• 23:00 - 23:30 thinks may be based on that my requirement and etcetera what sort of demand i am going like if we the popular example buildings to give is that ah a consider a car so buy or hm leasing a car may cost me something five hundred per day right whereas renting a car may be say for example five thousand per day ok so when it is economical suppose i by looking
• 23:30 - 24:00 at it is always the buying the car may be economical but i if i am commuting say large distance or say one scene that means one once in a month or couple of delayed days in a month then buying of car may be more costly than renting a car but if if i require that car on a daily basis that going to my workplace traveling a large distance etcetera then the
• 24:00 - 24:30 buy of car buying a car will maybe economical than renting a car all right so it all depends that what sort of demand demands you are having so you just to do some simple some mathematic expression ah little bit simplified so that try to have like suppose i have a demand dt so demand varies over time zero to capital t zero to time so demand for resources dt p is the max demand or peak demand ok a is the average demand so average over
• 24:30 - 25:00 the time scale b is the baseline or own unit cost so if i only unit cost what is the baseline cost c is the cloud unit cost so what is the cloud unit cost if i purchase the thing and you utility premium rent ah rent a car for that there is a some typo so there is a rent
• 25:00 - 25:30 that car maybe something it will come whatever in our case is five thousand by five hundred so something ten is the utility yeah so utility of the premium is that taking a cloud service divided by the baseline service right that is the utility premium we are having the things right that is the utility premium now so i have a variable demand dt i have a peak demand p we have a average demand a
• 25:30 - 26:00 there is a baseline owned unit cost dc cloud unit cost c and i want to find out the utility premium that is c by b now if i want to see the overall cloud cost so u into b into dt integrating zero to t is the overall cloud cost which is somewhat u abt is the overall
• 26:00 - 26:30 costing of my ah cloud if my if i want to calculate the overall ah baseline over time t so p that is the peak demand into because whenever i am going for based my baseline i have to go for the peak demand thing to calculate the thing b is the baseline own unit cost and t is the overall time scale now when cloud is cheaper when the ct is less than bt if the cost of cloud is less than
• 26:30 - 27:00 the cost of baseline then it is cheaper so or in other sense if you look at it very simplified form so when p by a is greater than the utility premium right peak cost by average cost a peak demand by average demand is greater than the utility premium so when the utility premium is less than the ratio of peak demand to average demand then your cloud may be ah cheaper then owning the infrastructure or owning the services so utility pricing in real world in praxis
• 27:00 - 27:30 demands are often offense pipe spiky like new stories suddenly it came into things market promotions product launches internet flash flood something goes to the internet etcetera some seasonal things like christmas tax so those time things goes up often hybrid model is based like you own a car for daily commute but rent a car when traveling or when you
• 27:30 - 28:00 need a larger move to a larger distance to move key factor is again the ratio of peak to average demand so key factor is there again that what is the ratio between the peak to average demand but we should also consider other cost like which had not considered in the previous calculation then our network cost both fixed and usage costs interoperability
• 28:00 - 28:30 overhead like different one information a one data is talking to other there in overhead consider reliability accessibility and so on so forth right so these are the different factors another aspect we will just quickly see the value of on demand services so simple problem when owning your own resources you pay the penalty whenever the resources do not match with the instantaneous demand suppose i have hundred resources and i pay the penalty when
• 28:30 - 29:00 it is underutilized suppose it is utilized with eighty eighty percent seventy percent then i pay the penalty for the rest of the ten to a twenty thirty things right either pay for unused resources or suffer penalty of missing service delivery things are there right so if it is ah or if it is higher things then i dont able to service so penalty is how do i calculate penalty is proportional to the demand instantaneous demand
• 29:00 - 29:30 minus instantaneous resources integrate over time t all right if demand is flat penalty is zero right if demand is linear periodic provisioning is acceptable right if it is a linear than periodic provisioning the acceptable if the demand is nonlinear then periodic provisioning in cloud is a big question right suppose if the demand is exponential like in this case ah dt equal to e to the power t right any fixed provisioning time interval tp according
• 29:30 - 30:00 to the current demand we will felt you will fall exponentially behind suppose i require time tp to provision the things right by the time i provision it has gone up so it goes on so this at t equal to e of t minus tp this time div that time to provisioning tp will create a havoc right so dt minus rt equal to k one dt if you say so penalty of cost
• 30:00 - 30:30 is c k one e to the power t in other sense this penalty grows exponentially it is extremely difficult to match this unless you over provision and try to look at that is also at times difficult because it grows in exponential things so if you need to be careful that when what type of demands we are expecting need to study and based on that the provisioning should be there so based on these i have i kept a small assignment for you to work at your at your own time and we will discuss this assignment
• 30:30 - 31:00 in one of these classes during some free time so it says that consider a peak computing demand of an organization one twenty units and demand of the functions time express at this like these are the demand of with respect dt is that demand over t the resource provisioning of the cloud to satisfied the current demand t is at t equal to so and so forth where delta
• 31:00 - 31:30 t is the delay in provisioning extra computing resource on demand so delta del this tilde is the delay right so what we say that the cost to provision unit cloud resource for unit time is point nine units so we want to calculate the penalty if any and draw the inference like why this type of penalty how things are there its so and so forth so we are assuming some of the factors like what sort of the what should be the delay in provisioning penalty it may
• 31:30 - 32:00 be either pay for on his users or missing service delivery and type of things so this i encourage you to look at this problem with this thing that while things will be there will in subsequent class one of this class we will discuss this problem so thats all for today so we looked at out the economy of cloud and where it is beneficial and what are the different aspect to which drives the economy of cloud and type of things
• 32:00 - 32:30 thank you