News Stay informed about the latest enterprise technology news and product updates.

Quan on the cloud, part 1: IBM Autonomics director heralds user facing applications

What does cloud computing mean at IBM? Dennis Quan, director of development in IBM's Autonomic Computing division, answers that question in this interview.

What does cloud computing mean at IBM? Dennis Quan, director of development in IBM's Autonomic Computing division, answers that question in this interview. Part 1 asks just what is this phenomenon called 'Cloud.' In Part 2, Quan will expand on the relation of Cloud Computing to SOA. Read part two.

How does IBM define Cloud Computing?
I usual describe it as a phenomenon from two different perspectives. One is the end user's perspective, the other is the IT perspective. What are end users going to be seeing out of the Cloud? Eric Schmidt, the CEO of Google, says in the future people's applications and information are going to be accessible to them from anywhere because it will all be up there in the Cloud. You won't have to worry about where it's being managed, how it's being stored. As long as you have access to the network, you have a connected mobile device, you'll be able to access your information. So it's a network based model that allows you to get to your services and information from anywhere. What about from an IT perspective?
From an IT perspective there are implications that go beyond what we traditionally thought of as the network computing model, especially given the kinds of applications users are coming to expect. Look at the Web 2.0 applications that are out there. That's really at the heart of the Cloud phenomenon. Look at YouTube and MySpace and Gmail. It's not just that they have lots of users coming on to these sites. It's also that these sites are harboring huge amounts of information. For a company to be able to make use of that kind of information for customer analytics, business intelligence, targeted advertising requires a different compute infrastructure than we were thinking about up to the Web 2.0 era. From the IT perspective we have to ask how do we create a scaleable infrastructure to support applications that deal with millions and maybe billions of users. There is a prediction that there will be four billion cell phones out there by 2010. And these users are going to be dealing with massive amounts of information. Sites are going to need process and handle and manage all that information. So is the Cloud the answer?
Cloud computing is about creating data centers that allow application providers to create Websites and services that can scale. People talk about the quality of elasticity with Cloud computing. On day one of a new application you might get a certain load of users. Then on the next day as the popularity of it grows, you might have a huge up-spike in the demand for that service. You want to be able tap into additional hardware resources. If you were managing a data center yourself you'd have to go buy machines. You might not be able to get them on line in time for the increase in hit rates for that application. Elasticity is the answer?
You tell the Cloud you need more resources and it finds a way to get those resources. The Cloud can expand or contract depending on the needs of your application at any point in time. That's an important piece of the Cloud's scalability model. Wasn't that the idea of the Grid?
Cloud computing is just a point in the evolution in the bigger picture of distributed computing that IBM has been involved in for decades, dating back to Grid technology and utility computing. With those models you did see the ability to perform large scalable compute tasks over a large number of machines. But the use cases tended to be scientific or mathematical, involving simulation of back office batch processing. So did the Cloud grow out of the Grid concept?
Yes. If you look at the way the Web 2.0 companies have built out Cloud technologies a lot of them evolved out of approaches we used for Grid. What makes the Cloud different?
Cloud computing is really driven by user facing applications where you need both large scale horsepower on the backend to do the processing, as well as the ability to run these large scaleable Websites that people interact with.

In part two of this interview, Quan discusses the relationship between Cloud computing and service-oriented architecture (SOA), data centers supporting the Cloud, as well as IBM's partnership with Google and the Android mobile platform.

Dig Deeper on Cloud APIs and integration

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.