Cloud Computing Overview: Part One

if I fall by the wayside yet again. But yes, I am here attending sweet World eighteen is a big cloud computing conference, and uh, I just wanted to start producing some episodes while I'm here. This is one of those special things where when I go to events and conferences, my goal is to give you guys more of the stuff you like. So I'm gonna drop a few episodes this week. Hopefully, uh you'll get them as the week goes on. There's going to be one today about cloud computing.

But I also plan on doing specific ones about net sweet That is the organization that throws Sweet World every year. An Oracle, which is net sweets parent company, Oracle purchased net Suite, a couple of years ago, and so I'm going to cover those as well as well as maybe any other kind of cool topics that I happened to encounter while I'm here at the conference. So far, it's been kind of an overview, So that's where I'm going with UH for now. And I've I've talked about cloud

computing in the past. Um I've I've covered it a couple of times on tech stuff, but it kind of bears going over again because it's one of those big buzzword terms that I think a lot of people have heard and most people, I think, by this point have a fairly good grasp of what it's all about. But there's still some confusion out there, and I kind of

want to clear that up. So we're gonna talk about what exactly is cloud computing, How did cloud computing evolve over time, because the idea is pretty old, why is it such a big business, and and how does it actually work? What's going on there? Well, first things first,

cloud computing is a very broad term. It covers an enormous number of different offerings and services, but in general, it's a method of delivering services over a network, essentially allowing people to access a computer that does not belong to them in order to do something. Now that something could be storing information, it could be processing information. It could be uh developing an app, it could be all sorts of stuff, and we'll cover a lot of that

in this episode. But your computer, the one that you are using to connect to the service, is not doing the work. In fact, it could be a very bare bones computer. That's the basis behind some of the very

weight notebook computers like chromebooks. The idea is that you really don't need to have a lot of bells and whistles in the laptop itself because you can access all the applications and services that you require using the Internet, and some other computer out there in the great, big scary world is doing all the work for you, so your computer doesn't have to. That is in some ways a big benefit to the end user because it can

really cut down on your upfront costs. Right because if you are just going out to buy a very simple computer, one that does not have an incredibly powerful processor or a whole lot of storage space or whatever, then the price can be lower. It doesn't have to be as big a ticket item for you when you go out there and purchase it. On the flip side, a lot of these services end up being subscription based services, so

you're paying to use the service in some way. Some of them are add based rather than fee based, so you're still technically generating revenue, but you're doing it by looking at ads as opposed to paying a subscription service. But in a long term, over a really long use case scenario, it could be argued that you're a little,

you know, lightweight chromebook kind of of computer. Cost more than if you had purchased a computer with native software stored on it, but you would have less capabilities as well. Now I'm going to cover all of this and more detailed throughout this episode, so just know the cloud computing it's a it's a pretty big deal. It's it's not just big as in it makes huge amounts of money. I mean, there's a multibillion dollar industry. But it's a big deal because it covers so much ground as far

as the features that can be grouped under a cloud computing. Now, the cloud computing model as you take advantage of another computer's capabilities to bolster your own, and like I said, the service could be as simple as cloud storage. That just means that you're storing files on some other computer systems somewhere. Typically that would be in some sort of huge data center, and normally that would mean the files were actually on multiple computers throughout the data center, not

just one, and that's for the purposes of redundancy. That way, if one of those computers should fail, you can still access your files. So if you were to visit one of these giant data centers, first of all, you would noticed that the air conditioning system is on overdrive to keep the computers cool enough to operate at maximum efficiency.

You would also notice just row upon row upon row of server stacks, and each of those servers would represent a device that could be processing information or storing information for customers and your own personal photos. Let's say that you're visiting the data center that belongs to the company that offers up an online photo album service that you use.

You would probably discover, if such a thing were possible, that your photos existed on multiple computers within that data center, and multiple copies of each photo would be on different uh would be within that data center, they would be on different servers, not multiple copies on the same server, but a copy might be on server A, and another one might be on server M, and another one might

be on server B seventeen, that kind of thing. And the whole purpose of that is if one of those machines should fail, you would be able to get your images anyway, because it's got this redundant system. If you don't use redundancy, you're pretty much playing chicken with fate. It's not a great idea anyway. You probably use some form of cloud storage fairly regularly, so it could be a photo album that's a very popular one that backs

up all images that you might take with your smartphone. Um, a lot of smartphones come with this just as part and parcel with the smartphone itself, where you have a setting where it's automatically uploading these images to a server bank somewhere so that you can access it wherever you are. Even if you were to change phones, you would still be able to access those images because they would be associated with your account, not with the hardware itself. Not

very common feature. Or you might use something like Google Drive, which allows you to upload files to Google servers or a very specific service within the Google Drive realm, the Google Suite realm, like Google Docs, where you can actually create documents or spreadsheets or presentations, all of which live on the cloud and not on your personal computer, unless you choose to download a copy, which you can do. Now, that last example gives a little hint of what else

cloud computing can do, because not just you can't. It's not just storing information. You can let you run an application or program on another computer, which means your own computer does not have to dedicate a lot of processing power to that job. Instead, all you have to do is run the program that connects you to that online service, and frequently that's something as simple as web browser, So your web browser becomes the interface with the programs you're using.

So with docs services like Google Docs or Microsoft Office three and the one drive Microsoft product, it's all about creating documents. But that's just the tip of the iceberg. There are companies that offer deep and complex services, some catering to specific industries or processes. Now I have a little bit more to say about the history of cloud computing, but first let's take a quick break to thank our sponsor.

The power of cloud computing is that it gives users a chance to leverage computing power they might otherwise never have access to. So for businesses, it can mean being able to scale processes smoothly while the business grows. So instead of spending a lot of money buying powerful computers or setting up your own data centers, the business can become customers of one or more cloud computing service operators and offload all of that work to a different company

and thus scale up faster. Because you know, every time you would grow as a company, there would be kind of the stop start situation where you would need to build out your infrastructure to make it larger and more robust to deal with all that growth, But then you would end up filling that right. If your business continues to grow, you would eventually hit that that limit that your new infrastructure can handle, and then you have to

do it all over again. And this is a very complex process at times, like if your business gets really really big, with enormous departments underneath that business, each of those departments having its own needs, then it quickly can become unmanageable unless you have really got your act together. So what these services tend to do is make it

much easier for companies to scale as they grow. This is particularly important in today's business world where you've got all these startup companies, a lot of which are begun by people who are first time business owners. Right. These are entrepreneurs who came up with an idea. It was a really good idea, it resonates with their customers, they launch it, it becomes way more popular than they were

necessarily prepared to handle at the beginning. They grow very rapidly, and that can be the doom of a small company. A small company can actually collapse because it was growing so fast that uncontrolled growth can mean that people make bad decisions about where they need to put their money in order to sustain the company as well as to continue to grow. So these services are what allow companies like that to kind of not worry so much about

that part of their business. It can be put into the hands of another company that specializes in those UH processes, which frequently can be customized to the specific customer, so it's not like a one size fits all approach. But the whole point of it is that the business owners don't have to worry about all that themselves. They can actually work and partner with these other companies to handle some of that while they the business owners continue to

concentrate on the very core of the business itself. So that's why cloud computing has become such a huge deal. It is an enabler for other companies to grow. Not to mention, it is also a very useful tool for the average person. I use cloud computing all the time.

I use it for creating my documents at work, but I also use it for my home stuff, Like if I want to do a personal project, I use cloud computing for that because I can access that information wherever I am by logging into the associated account and then using whatever computer I'm on to access that information. Very very convenient. Uh. But that's just a high level overview of what cloud computing is. And there are some questions that come up, or at least should come up, when

you hear about this model. Among them are questions of ownership, because if your documents, files, and processes live on someone else's machines, who owns that stuff. There are a lot of confusing terms of service out there that seem to suggest your handing over ownership of your stuff to these third parties, but that can be a little a little misleading. I'll cover that a little bit more later on. But there are other questions too. If the service were to

go out of business. Let's say that you have relied upon a service that's holding up your back end operations of your business. If that service goes out of this, what happens to your organization? If it's cloud storage to your files go away forever? Would you end up collapsing

right behind the service provider? Because that would become a domino effect, right if a major cloud service operator were to run into serious financial problems, maybe there's a terrible scandal, maybe there's a data breach, maybe something happens that just affects that company at a fundamental level. What does that do to the customers? Does that does it cripple multiple other businesses? Could you see an entire, attire swath of companies collapse as a result of a provider collapsing? And

then there's the issue of security itself. Can you, as a customer of a cloud computing service operator be sure that your information, that your processes, that your infrastructure in some cases are safe from bad actors? After all, if the cloud computing operator is a really big one, it's

probably a really attractive target for hackers. I'll talk more about some of those questions a bit later, But first I want to talk about the evolution of cloud computing itself, because it goes back a bit further than a lot

of people think. The term cloud computing for the average person became kind of a big buzzword around two thousand nine, two thousand ten or so, and when it first popped up in general use, it caused some confusion, so much so that I actually got an assignment from how Stuff Works to write more than a dozen articles for the website at that time, and it was covering stuff like cloud computing, grid computing, utility computing, and related terms to

help clarify things. Those are the days now. The actual term cloud computing goes back more than a decade before it became a buzzword on the average streets of the world, right like before the average person was aware of what cloud computing think was, the term was already in use for more than a decade. But the concept of cloud

computing is even older than that. So to understand the birth of cloud computing, we actually have to go back before the era of personal computers, even before the era where businesses, big businesses, we're using computers for the average employee. We're going all the way back to the nineteen fifties. Not at that time computers were mainframe computers. They were giant, expensive machines that took up the better part of a room, sometimes the better part of a floor of a building.

This would not be practical to have as a computer that one person would use, right, it's so big and if you wanted to have multiple people use a computer, and you need to have multiple computers, then you would be dedicating floors of your buildings to just the just a few computers. It's it was not an efficient means. And in fact, that was one of the big problems

with the computers early on. It wasn't there uh their processing power, or their storage capacity or anything that The problem was, how do you make this powerful machine efficient? How do you best make use of its capabilities? Because if it's so large and if only one person can use it at a time, then you've got limitations on what the computer can do, just based upon human limitations.

There's only so much a human can do with a computer like that, and it meant that there were very few use cases for a computer that were compelling enough to to really uh justify the expense of one. The solution in the nineties sixties was to create computer terminals, which typically consisted of a workstation that has like a keyboard and maybe some other input devices, and then an output device like a display. The terminal wasn't in of itself a computer, It was more of an interface for

accessing the main frame computer. So you still have a centralized computer, but then you would have one or more term annals that would allow users to interface with the computer, and then, when paired with a strategy called time sharing,

a main frame could support multiple terminals. This concept was first described by John Backus in nineteen fifty four, but it took several years for someone to come up with a practical means of enabling it, and in nineteen sixty one there was a computer scientist and the father of artificial intelligence, John McCarthy, who proposed that the time sharing model, once fully developed, could give rise to the shift of

computer as a utility service. So, in other words, in the future, he would argue, customers could pay to make use of a computer's processing power and for specific applications as well, so instead of owning a computer, you would pay someone else to make use of their computer's abilities to complete a task. Uh. Now, this did not first see the development of the personal computer, because again, at the time, computers were enormous, They were very complicated, they

were difficult to work with. You had to have a very specialized set of knowledge and skills in order to work with these computers. So it seemed like it was well outside the realm of the average person. People weren't thinking in terms of personal computers. So what McCarthy was proposing was a model where you would still be able to take advantage of a computer's processing power, but you would pay on demand. Right, You would pay whenever you needed to use a computer, and then in return you

would get some application run or some computation performed. And this would actually become part of the foundation for modern cloud computing. While we kind of bypassed this for a while, it's come back around. Utility computing is related to cloud computing. It's not synonymous, but it is related to it. I've got a little bit more to say about cloud computing, but first let's take another quick break to thank our sponsor.

The development of multi user mainframe solutions was originally funded by the Advanced Research Projects Agency or ARPA, the predecessor to DARPA, the research and development arm of the United States Department of Defense. They've been responsible for funding tons of projects that have been all about pushing technology forward, frequently with at least a a an eye toward defense, but not necessarily all about that. So ARPA NET is a great example, and I'll talk about that a little

bit more in just a minute. But ARPA or DARPA as it now is known it changed its name in the early nineteen seventies. DARPA has also funded big projects that have led to things like robots that can respond to various uh crises, like there was a big robotics challenge where the robots had to perform a series of of different tasks in order to simulate responding to a nuclear meltdown scenario, or driverless cars. DARPA has also funded

a lot of the development behind autonomous vehicles. So back in ninety three, the agency ARPA awarded two million dollars to m I T to take on what was called Project MAC, which was also known as the Project of Mathematics and Computation Now. Later on, the acronym was tweaked a few different times to mean various things, one of which was multiple access computer, which was one aspect of

Project MAC, but not the only one. The ARPA side of the project was headed by a guy named Joseph Carl Robnett Licklider, better known as j. C. R or Lick, and he had previously served as a professor at m I T before joining the Department of Defense. The m I T side was originally headed by Robert fano A. Project MAC would span many different areas of computer science, but the one we're specifically interested in here is time sharing.

The origins of the time sharing model came out of a different project and m I T, but was completed as part of Project MAC. Lick Lighter is going to pop up again in our discussion in just a second now. Typically, time sharing meant that the main frame would devote a certain number of processing cycles to each terminal's operations, so you could have multiple people using different terminals accessing the same mainframe computer, and the main frame would work on

problems in sequence rather than simultaneously. But computers work very quickly, and we humans are relatively slow by comparisons, so on casual glance, it looks like one computer is handling all the work of all the terminals all at the same time. In fact, the reason time sharing works so well is that we humans tend to work in short batches or spurts.

This means that the main frame would be waiting on humans more than humans would wait on the main frame, which is another reason why one person working on a computer is not an efficient use of that computer, because there's a lot of downtime when you look at the accumulation of all the little pauses and breaks humans take. So if you could allow multiple people to access the same machine at the same time, you can then shift the downtime of one user so it becomes the active

time of another user. And with the right number of users, that would maximize the efficiency of the main frame itself. It would essentially be working constantly, and in turn that would justify the massive cost of owning and operating a mainframe computer. Time sharing also opened up the possibility of allowing a business to make use of a computer owned

by someone else. The business would just have to have access to a terminal for the main frame in question, so you could argue that's the very foundation of the concept of cloud computing. Lick Lighter is one of the

people responsible for the formation of the Arpanett project. Ar Ponnet's goal was to create means for different computers and different locations to share information with one another and therefore resources between each other, even if the computers were of different make and model, which was not a trivial task as different mainframes relied on very different computer architectures and different operating systems, which made them incompatible with one another

in normal use, so it's like they're speaking totally different languages. Lig Lighter envisioned what he called an intergalactic computer network, and in this vision, link Letters saw everyone in the world having access to computers and information through an interconnected network of machines. Essentially, he was proposing the Internet. The Arpanet project launched in nineteen sixty nine. In the early nineteen seventies, another big development took place, which was the

evolution of machine virtualization. In the nineteen sixties, we use the term virtualization to refer to the ability to allow multiple users to access the same machine more or less sim sultaneously, as it was almost as if each user had access to his or her own virtual computer. So essentially it was another kind of word for time sharing, But in the nineties seventies the meaning began to shift as computer scientists found ways to create virtual machines within

a computer itself. The computer runs software that allows it to partition off assets to run an instance, as if it had a second machine inside the main machine. It takes up some of the primary machines abilities, some of its processing abilities, some of its storage. It's it's almost like you've just divided one main frame into two computers, but you're doing it on the software level, not the hardware level, and you could even run different operating systems

on the same computer. This way, you could create essentially what amounted to an emulator that would allow software mint for a different type of computer to run on another computer. We do this today. There are a lot of computers out there that run uh different operating systems simultaneously, and

you can switch back and forth. And it's essentially this virtualization software that allows the computer to divide itself up so that some of its assets are dedicated to running one operating system and the other assets are dedicated to running another operating system. So you might have Linux and Windows for example, or Max had virtualization that allows you

to run Windows on a Mac. So that way you could run Windows based software on the Windows operating system platform as if it were a Windows PC, or you could switch over to the Max side and run Mac software that was designed to run on the Mac operating system. Because remember, different programs meant for specific operating systems are not interoperable with other operating systems. So virtualization began to take on a slightly different word, different meaning rather in

the nineteen seventies. But it also meant that you could create a virtual machine that someone could access, and that's another one of those little founding notions of cloud computing. Technology would continue to evolve. Our bonnet would pave the way for the Internet, which sort of began to coalesce in the late seventies early eighties. Uh. The Web, of course, would be another evolution of that. You would start, you know,

a layer on top of the Internet. The Web and the Internet are not synonymous, they're not the same thing. But the Web is a layer that exists on top of the Internet, and it's frequently the way most of us access the Internet, and we tend to access it using the web and email. Those are two big ways. And then we also have apps which can access the Internet. They often have a web like interface, so a lot of times we associate that with the Web, but they

could be completely independent of the Web. So we also access the Internet through apps, but I think most us kind of associate the Web with the Internet as being almost one and the same, although you should keep in mind that's not exactly true. Anyway, people before the Web would call into other computers using modems, uh, and they would upload or download files from them. They could do

that on bulletin board systems. I've talked about those recently, but then also later on through Internet servers, you could use things like tell net to log in remotely into another machine. Guys have got a lot more I want to say about cloud computing, but in order to get these episodes out properly, I'm going to cut this one a little short because Hotel WiFi makes it really hard

to upload larger sound files. So we're gonna rejoin this conversation with a discussion about some other forms of computing that are similar to cloud computing, like grid computing, and then we'll talk a little bit about what cloud computing is used for today. Some of the more recent developments in the world and uh then we'll conclude before we move on to our next subject. So I want to thank everyone out there for listening to this bonus episode

of text Stuff. If you have suggestions for future episode topics, let me know. Send me an email the addresses tech stuff at how stuff works dot com, or drop me a line on Facebook or Twitter. The handle of both of those is tech stuff hs W. Remember you can follow us on Instagram and also go to twitch dot tv slash tech stuff to watch me record these shows live, at least the normal ones when I'm back in the studio. On these remote recordings, I don't tend to stream them

because said hotel WiFi problems. But I look forward to see you guys in the chat room and I'll talk to you again really soon for more on this and thousands of other topics. Is it how stuff works dot com.