Andre Charbonneau, grid computing and applications specialist at the National Research Council, recently won a CANARIE iWay award for new technology development, along with teammate colleague Gabriel Maeescu.
Charbonneau, who
developed the software that allows researchers to remotely access through a Web portal Canada’s largest nuclear magnetic resonance tool, which is housed at NRC’s Ottawa campus, spoke to ITBusiness.ca about his work on the project and what it means to scientists in the field.
ITBusiness.ca: Tell us about your role in this project.
Andre Charbonneau: I was the main developer, actually writing the code, and I also interacted with the users from various research departments such as biology and chemistry. I was running back and forth between the developer and the users.
ITB: What is SpectroGrid exactly?
AC: SpectroGrid is two components: one is for scheduling and booking of the instruments. The other is for remote instrumentation, which is data acquisition and data visualization.
ITB: Why is NRC doing this project?
AC: NMR (nuclear magnetic resonance) systems are really important systems, but unfortunately, they are really expensive systems. Before the NMR project systems were located in the office of the researchers who were using them. But recently they were all relocated to a building at the NRC and a new machine is going to arrive soon. It’s going to be the biggest in Canada. The users will not always be able to work right near the machines so they will have to be able to work remotely. So this SpectroGrid project will allow users to easily access the system in a secure manner remotely and there will be an easy-to-use Web interface so they can schedule when they’re going to use the instrument and make reservations so they don’t conflict with each other.
ITB: What were some of the challenges in developing this application?
AC: The one challenge that comes up often is making the users use the system because if you don’t have any users you don’t have a system so it’s making the users happy; that was our main goal. That means packaging the software in a way that is easy to install on the user’s system. If it becomes too complicated, users will simply stop using it and then the project goes nowhere. It also implies writing a lot of documentation that is easy to understand for researchers, who are not necessarily computer programmers, so that meant a lot of interaction between me and the researchers, because I don’t know about chemistry and biology. Also another challenge was some of the systems were running legacy software and we had to adapt our technology to be able to work with those legacy systems as much as possible. It turned out for some of those systems the processor was just too slow, and we simply had to either replace the software with newer versions or if that wasn’t possible wait for a new system to arrive later.
ITB: What security was this application built on?
AC: Right now we’re using the grid security infrastructure. That means every user of the system will have his own certificate that authenticates that user onto the system. That’s a really secure system everything is encrypted. No password is sent as clear text, so that will protect the user’s data. Also when he does online scheduling through the Web portal he also authenticates using the same grid certificate.
ITB: What is the application for this research?
AC: From what I’ve seen it has lot of application in chemistry and biology. Let’s say you have a solid and you want to know what kind of molecules are in the solid or the concentration you can use NMR.
ITB: It sounds like the kind of research that has to be done in certain sequences, rather than having many scientists doing research at the same time on different aspects.
AC: The way it works is every user of the SpectroGrid system has a role. Some of them have administrative privileges and those people are able to create projects and find users and also every user is only allowed to use either an NMR system or another and also a certain amount of time a month. There’s a quota. The way it works is the user makes a reservation that he would like to make an experiment on that date and there will be someone monitoring the system and looking on a daily basis who is going to use the system and prepare the instrument for that experiment. It involves putting in some solids or liquids in the machines at the time the user requests.
ITB: So the time management component is not automated.
AC: No, users have to make their reservations. The system won’t make an optimal reservation for the users, because there’s a lot of human interaction required to put the samples in the machines. That’s why the users have to book themselves and there will also be some interaction between the NMR administrator and the user through e-mail or phone.
ITB: What happens if they have problems using the system?
AC: Currently if they have a problem it’s all controlled through the NMR software, so it’s the same as if they were running it locally. If there’s a problem with the system they shut down the software remotely and then each researcher has an e-mail address or phone number to call the administrator onsite. On the software side of things if they find some problems with the SpectroGrid infrastructure there’s a mailing list where they can log reports or they can e-mail me directly and we can take care of the problem right away.
ITB: When do you expect an update to the release and what kinds of capabilities do you expect to add eventually?
AC: Right now it’s in production release. People are using it on a daily basis but there are already new features that are planned for future releases. They come in on a monthly basis, so a lot of features still need to be implemented but the basic stuff is there right now. The architecture is made so it can be applied to other instruments other than NMR, such as electronic microscopes or accelerators, so it’s easy to adapt to other instruments.
ITB: will there be any problems with availability of the network? How would that affect the system?
AC: We did some tests through the CANARIE high-performance network and we’re very pleased. A lot of the time the users don’t even see a difference in the response when they use the software. We also tested the network from Alberta and the response was great too. We had some problems before but it was not related to the network itself, it was related to a system. The processor on the system was too slow to handle the encryption that goes on when you use the secure GSI, but so far the network has been good.
ITB: what does winning the award do for you?
AC: It means a lot to me and to the NRC as a whole. NRC has always been one of leaders in leading-edge research and winning an award like that is proof we’re going in the right direction, especially coming from an organization like CANARIE.
Comment: info@itbusiness.ca