Will Accessibility Rain on Your Cloud Computing Parade (1 of 4)
Articles,  Blog

Will Accessibility Rain on Your Cloud Computing Parade (1 of 4)


Hello and welcome to the session titled “Will Accessibility Rain on Your Cloud Computing Parade?” My name is Elizabeth Woodward and I’ll be presenting to you today. I’m a senior software engineer and an inventor working with IBM Research’s Human Ability and Accessibility Center and our mission is to make sure that our software and hardware and anything really that we deliver is accessible and usable by people of all abilities. The way that I became involved in cloud computing was that I saw this growing wonderful opportunity to be able to leverage cloud computing to provide a variety of personalization services to people of all abilities and to meet people’s preferences. So things like being able to provide a service to convert the literacy level… to being able to the expand upon text to speech and speech to text services that we currently see available on the cloud and, in doing that, I had a chance to work directly with some cloud infrastructure deployments and I had a chance to encounter some inaccessible solutions and some things that really caused problems for people of different abilities. So that has really spurred my interests and provided a background. And, in addition to that, I have been working with our organization to explore, define new delivery models for the technologies that we create within Research. So we’re very closely working with teams on deploying our technologies as software is a service and as platform as a service, so I have some experiences that I’ll be sharing with you in those areas as well today. For our agenda today we’re going to start with the definition of what cloud computing is just so they were all on the same page. There are three different sections that we’re going to discuss. We’re going to talk about some wonderful opportunities that cloud computing is providing for accessibility and along with that we’ll talk about some of the challenges that we’re seeing some things that are causing us to not be able to fully take advantage of those cloud opportunities. And then finally we’re going to talk about the yellow umbrella that you saw in the previous slide and that yellow umbrella is that thing that’s going to keep accessibility from raining on your cloud computing parade. Let’s get started! There are essentially five characteristics that we’ll talk about here: broad network access, rapid elasticity measured service, on demand self service, and resource pooling. Let’s start with the resource pooling. What that means is that we’re going to bring a variety of resources together and make those available to multiple consumers. And these are going to be different physical resources that could be CPU or disk space for virtual resources. These might be access to virtual machines, virtual environments that have a complete set of software on them. And we’re going to make this available so they can be assigned and reassigned according to consumer demand there is a sense of location independence with resource pooling meaning that you don’t know exactly where those resources reside. All you know is that you have access to them. A second thing that you should know is that cloud computing provides broad network access. What this means is that we are going to make capabilities available over the network and you can access them through your standard mechanisms and they’re accessible by a variety of platforms: different mobile phones, tablets, laptops, workstations will have broad network access. With a computing we also have rapid elasticity. What this means is that the capabilities that we’ve been describing they can be elastically provisioned and released. That means in the past we had to have the resources of web servers available that could always provide the maximum amount of resources or access that we expected at any given time. So, for example, if at Christmas time you’re expecting for there to be a huge load on your servers because people are going to be just buying like crazy, then you’re going to want to make sure that at all times or at least during the Christmas season that you have configured your servers to have the right amount of space and that you have the right number of servers all working together to provide that. Cloud computing takes some of the thinking out of that. What that means is that if at any moment we have a large numbers people will be accessing we immediately expand our resources to be able to serve them and then when they’re finished when they go away after the holidays we can reduce the services that are provided specifically for that purpose. So you can see that there’s this opportunity in there to be able to charge by the sip, by the access, to expand and contract and charge for exactly what is being used rather than charging a set amount for a set period of time regardless of what the actual needs are. And that’s what we’re talking about with measured service–being able to measure the amount of resource that is actually being used and meeting the capabilities that are required with that. So there is, with measured service, there’s a level of an abstraction that’s appropriate to the service. For example, if I’m a using storage I might charge for the number of gigabytes or terabytes that are being used. If I am serving a number of users, I might just go ahead and charge for the number of active user accounts or bandwidth or processing. Regardless of the measurement, it is appropriate to what is being used and the consumer is being charged just for what is being used. On demand self service is another of the essential characteristics and what that means is that consumers can– on their own without the interaction of humans and the service provider–provision their own computing capabilities and this is standard, its automatic. They can increase or they can add use of server time and network usage again without having to have the service provider do that for them, interact with them, or perform those tests on their behalf. So now will turn our attention over to the four deployment models.We have the public, private, hybrid and community deployment models. We’ll start with the private cloud. With the private cloud, that means that the cloud is for exclusive use by an organization. For example, IBM has a private cloud or IBM Research has a private cloud. UT–the University of Texas could have their own private cloud. This is a cloud that’s intended for use by a single organization and there can be multiple consumers in there. You can have multiple business units for example but this is considered one homogeneous cloud environment. If we take a step beyond that there are community clouds and this community cloud concept means that were providing a group of communities and consumers with access to a single cloud infrastructure. These groups might have the same mission, or security requirements or policy or compliance. But they are very closely linked and tied and working together For example, we previously said that the University of Texas could have their own private cloud. For the community cloud we might extend that to include all of the University of Texas universities. A public cloud, that’s provisioned for open use by the general public and that generally resides on the premises of the cloud provider. For private and community we didn’t really talk about where the equipment resides but generally it can either reside on the premises of the target consumer or the cloud provider on can provide access on their own hardware and systems. With public cloud that infrastructure is available on the premises of the cloud provider Then there is the idea of a hybrid cloud. With a hybrid cloud we have two or more of these distinct cloud infrastructures: private, public or community and they still have their unique aspects that are bound together and working together with some standards or proprietary technology that enables data and applications to be portable and ported across a cloud infrastructure. So they are in harmony and they’re working together, but you have these different varieties. So, for example, I might have a private cloud that accesses the services available on a public cloud and there might be some interaction between the two to customize or tailor that to the needs of my of organization that’s working in the private cloud but there is this connection between the two. So now that we have a common understanding of what we mean by cloud, I’m not going to go into a definition of accessibility right now because we’ll be talking about accessibility throughout this entire presentation

Leave a Reply

Your email address will not be published. Required fields are marked *