The types of computing service, these are the main acronyms that you will think of or when you're discussing Cloud, you'll see this term IaaS or Infrastructure-as-a-Service, where it's basically the most basic form of a Cloud compute or computing Cloud. It gives users access to infrastructure basics such as a server, or storage, or some sort of networking appliance with they would use. This model comes the closest to replicating the functionality of essentially a standard network that you would deploy at a normal datacenter on premise. The Platform-as-a-Service is a PaaS, and this takes the infrastructure one step further, and offering that is a complete development environment with the operating system, servers, tools, and programming already in place. Rather than just renting racks of equipment, you're actually renting an operating system. You're getting a server that's already been licensed and installed and ready to go. Again, pay-as-you-go gives you a little more scalability. Then another good defining term is SaaS, or Software-as-a-Service. Where instead of the servers and then installing an operating system or buying the servers with the operating system installed and then installing your applications, you could actually just spin up the application. That application is already ready to go. No need to install the underlying infrastructure or apps themselves, and this would be like a typical business that like CRM for the enterprise, or big data analytics, or maybe a video hosting platform, or a voice call engine, or communications engine, or something like that. These would all be services that are hosted and you would pay for access to them, but you wouldn't have to install or a manage them yourself. Cloud computing models. The cost of a Cloud computing model is available in two basic forms. You have the public Cloud and a private Cloud. The public Cloud is a shared Cloud infrastructure provider owned and managed. The benefit is you can you can rent as you need it. You lease the equipment as you need it, and when you don't need it, you relinquish it and you stop paying for it. Very inexpensive in the short-term, it can be a little more pricey rather than owning the gear, but you're saving a huge amount of capital expenditure even though you're operating expenditure might tend to creep up a bit. Private Cloud are firewall behind the enterprise Intranet or on-premise datacenter. The benefits of owning the Cloud Is like owning a car. You have the car, you can use it as much as you want to use it, you can fully configure it, manage it, it's your car. Whereas public Cloud would be like renting an Uber, or renting a car from a company where you only use the car when you need to go somewhere. A Hybrid Cloud would be a blend of the two. We have some instances that you use, the on-premise Cloud, your own servers, and other instances where it's more beneficial to use the Cloud. This might be a difference between, say, internal proprietary applications that you have to keep full control of, and there might be medical information that HIPAA regulations require you to keep on-premise. You might have government regulations like in Canada where there are laws in place to keep that data from ever crossing country boundaries into the United States. Those applications would be private, but other applications where security isn't as much as a concern, or privacy isn't as much of a concern, like say your Gmail or your Dropbox corporate access box folder storage thing, that could be on the public side. Hybrid Cloud is some applications go to servers on-premise, other applications are hosted in the Cloud. The public Cloud method is exactly as we said. No upfront investment, you've subscription-based, very flexible pricing based on service level agreements, fully managed service and focuses on quick growth, meetings scale, meeting demand, it's consumption model effectively. Dynamic growing computing needs, and your concerns with hosting something with public Cloud are, like I mentioned in the examples, security compliance, and of course, your paper use operational costs. For the large servers. Like if you were to install ClearPass. ClearPass is a beast of a server. The cost of hosting ClearPass in the Cloud would quickly outweigh the cost benefit so unless there's just something you're messing around with for a few days, it would usually install a large server like that on-premise, just simply give it the amount of RAM, and storage, and processing that goes into a high-performance server like that. With private Cloud, bringing that on-premise has a bunch of benefits. It's a dedicated Cloud or datacenter for typically a single organization although you can have additional tenants internally. The business units perceive multi-tenancy, but that's again, your business units, so maybe different departments have access to it. Higher initial costs, you have to buy the hardware, you have to install it, you have to wait for the delivery, you have to license it, and do the configuration. Then once it's built, and it's up, and it's ready to go, which takes a lot longer than just spinning it up in the Cloud, it is now on-premise and you own it. A private Cloud may be located on-premise or even off-premise if hosted by a service provider. In on-premise private Cloud, data is of course, stored at that site, which can make it easier to control your compliance issues. Cloud management skills could be a bit of a concern here where you just don't have familiarity with something like Amazon Web Services or Azure from Microsoft, and so you might be hosting these services on-premise in a more traditional method. It also minimizes your exposure to risk, your availability, your compliance, your scalability, your updates, that you do have to manage yourself. While you're never going exposure, exposure might be minimized. The fact that you have to maintain your up time, you have to maintain compliance on that, you have to maintain updates. A lot of those could be hosted through a PaaS or a software as a service provider. Then keeping up with fast-paced technology before you know it, your server's 18 months have gone by and now it's old news, whereas if you were just renting space on a rack, they would be managing those Rack mount servers for you. The Hybrid Cloud Model, you might have a corporate site here with at least a private and a public Cloud sided to the architecture, so Corp Site 1, Corp Site 2, and while they have their own local site storage, they might have provider-based solutions that are Cloud, and give them the availability to access that across different geographical locations. Hybrid Cloud can deliver the best of both the private and a public Cloud infrastructure, and applications can be partitioned to preside in both the public and private Cloud partitions. Even the same application where initially you use whatever scale you have on-premise, and then as you need to scale up additional servers, you could program that to spin up on-demand through a provider. What drives work load placement. In a hybrid environment, you have to decide which applications are going to be part of a workload. The main factors are user experience. If I have users that are spread all over the world, I would definitely benefit from a provider that could have servers in that continent or in that region of the world, and provide a much faster response time, rather than having to force all user traffic to find his way back to my main datacenter. Also your cost, your scalability. We've talked a lot about your security, and compliance, and legislation. There may be certain servers like the Canadian example that you don't want them to leave your country. If you have a big data solution, with a data might be placed in the Cloud for economic reasons, but the user interface might be put as close to the user as possible. Meaning on-premise just to reduce latency that way. Another situation for a database customer, the user might be a sales representative that are not at the office but are using mobile devices. Maybe the UI would benefit more being in the Cloud and these would be users all across the world, all across your country and the database might be kept on-premise for security reasons or for those concerns. This is not only true for applications on-premise, but for services and activities at very edge by networks. We want to always consider where is the user? Where do we need to store the data? Are there certain security or compliance requirements that are in place here? Are their costs that are driving this placement? Those all should be taken into account.
