Service-oriented science is the future, says expat

Grid computing is changing the way scientists work

With access to KAREN (the Kiwi Advanced Research and Education Network), there is no excuse for local researchers not to turn up on the world stage, says one of the fathers of grid computing, Kiwi expat Ian Foster.

Grid computing has changed the way we think about problems, and it allows us to do things that we previously never thought of doing, Foster said when speaking at KAREN’s very first event, held in Auckland last week.

Foster, who is professor at Argonne National Laboratories and the University of Chicago, cites initiatives such as the fMRI Data Centre, which aims to create an environment that will benefit the whole fMRI (functional magnetic resonance imaging used in brain activity scans) research community by providing access to a common data set that everyone can use in order to develop and evaluate methods, confirm hypotheses, and perform meta-analyses.

Through an online database with data mining capabilities, the fMRI Data Centre provides a public repository of peer-reviewed fMRI studies and all data necessary to interpret, analyse, and replicate these.

Grid technology is making it possible to instantly make data or analyses available to research colleagues, instead of having to download data from ftp-sites or waiting until the research is published in paper, says Foster.

While the first generation grid computing was about on-demand access and batch computing, the second generation paves the way for service-oriented science, says Foster. In his view, this approach is the future of e-research and it will also reduce time spent on mundane tasks.

Service-oriented architectures allow developers to provide information tools as services that clients can access, says Foster. This makes tools available to anyone, and previously manual data-processing and analysis tasks can be automated by having services access services, he says.

Service-oriented science is already being applied in the researcher community to a certain extent, but a lot more effort is required if this approach is going to become mainstream, he says. In order to create a service-oriented ecosystem people need to create services which others discover and may use to create a new function, or a mash-up, that will in turn be published as a new service, he says.

“But we need more than this; we also need someone to host the services, because I may not want to host services on my PC, for various reasons. So I find someone else to host my services — in the US it could be Amazon, or here in New Zealand, BeSTGRID [Broadband enabled Science and Technology GRID],” he says.

Grid technologies can accelerate the development and adoption of service-oriented science, he says. Foster also points out that grids are communities of people as well as computers, and these communities are based on trust. Communities may share computers, data, resources, sensor networks and services, and KAREN enables New Zealand researchers to be part of these communities, he says.

“You have the network, connecting you to the world at 622Mbit/s. All you need to do now is show up,” Foster says.

Many researchers feel that New Zealand is too far away to participate on the world stage, but hopefully that will change now, he says.

“But this is also a question of funding,” he says. “In the US you are funded to go to meetings, maybe we need something like that in New Zealand.”

Join the newsletter!

Error: Please check your email address.

Tags researcherservice-oriented architectureGrid ComputingKAREN

Show Comments